{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9999398785546805,
  "eval_steps": 500,
  "global_step": 8316,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "auxiliary_loss_clip": 0.04592296,
      "auxiliary_loss_mlp": 0.0257779,
      "balance_loss_clip": 2.47145319,
      "balance_loss_mlp": 2.09008121,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.30235029022519,
      "language_loss": 2.5798173,
      "learning_rate": 0.0,
      "loss": 1.90189219,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 15.610957145690918
    },
    {
      "auxiliary_loss_clip": 0.03096269,
      "auxiliary_loss_mlp": 0.01643611,
      "balance_loss_clip": 1.65457797,
      "balance_loss_mlp": 1.3195529,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 55.71490220387992,
      "language_loss": 1.89131498,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.93871379,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.637425184249878
    },
    {
      "auxiliary_loss_clip": 0.03069437,
      "auxiliary_loss_mlp": 0.01671734,
      "balance_loss_clip": 1.65279078,
      "balance_loss_mlp": 1.34996462,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 41.55070520941426,
      "language_loss": 1.61651754,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.66392922,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.4471964836120605
    },
    {
      "auxiliary_loss_clip": 0.03077628,
      "auxiliary_loss_mlp": 0.01737237,
      "balance_loss_clip": 1.65168333,
      "balance_loss_mlp": 1.41413188,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 36.590161607148275,
      "language_loss": 1.6432445,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.69139314,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.52978777885437
    },
    {
      "auxiliary_loss_clip": 0.03073762,
      "auxiliary_loss_mlp": 0.01667402,
      "balance_loss_clip": 1.6533705,
      "balance_loss_mlp": 1.35040045,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 55.09868365501697,
      "language_loss": 1.93286073,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.98027241,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.70761775970459
    },
    {
      "auxiliary_loss_clip": 0.03084434,
      "auxiliary_loss_mlp": 0.01683669,
      "balance_loss_clip": 1.65810478,
      "balance_loss_mlp": 1.36189914,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 46.660183596256495,
      "language_loss": 1.60990548,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65758657,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.817614793777466
    },
    {
      "auxiliary_loss_clip": 0.03136291,
      "auxiliary_loss_mlp": 0.01618009,
      "balance_loss_clip": 1.79442191,
      "balance_loss_mlp": 1.41659284,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.601151396540847,
      "language_loss": 0.81463128,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86217427,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.216681480407715
    },
    {
      "auxiliary_loss_clip": 0.03049894,
      "auxiliary_loss_mlp": 0.01707844,
      "balance_loss_clip": 1.64216876,
      "balance_loss_mlp": 1.39522922,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 40.64089087448784,
      "language_loss": 1.57914817,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.62672544,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.725752353668213
    },
    {
      "auxiliary_loss_clip": 0.03056056,
      "auxiliary_loss_mlp": 0.01697079,
      "balance_loss_clip": 1.64996397,
      "balance_loss_mlp": 1.37893343,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.814830989409774,
      "language_loss": 1.47760499,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52513647,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.7891881465911865
    },
    {
      "auxiliary_loss_clip": 0.03042627,
      "auxiliary_loss_mlp": 0.01655872,
      "balance_loss_clip": 1.6488142,
      "balance_loss_mlp": 1.3390615,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.224072286565876,
      "language_loss": 1.52933049,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57631564,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.8020663261413574
    },
    {
      "auxiliary_loss_clip": 0.0305074,
      "auxiliary_loss_mlp": 0.01682859,
      "balance_loss_clip": 1.64492202,
      "balance_loss_mlp": 1.3784461,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 14.579988724890498,
      "language_loss": 1.32469368,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.37202978,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.880160093307495
    },
    {
      "auxiliary_loss_clip": 0.03064584,
      "auxiliary_loss_mlp": 0.01705995,
      "balance_loss_clip": 1.649616,
      "balance_loss_mlp": 1.37812221,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 29.521131447989053,
      "language_loss": 1.25251997,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.30022573,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.798915386199951
    },
    {
      "auxiliary_loss_clip": 0.03045907,
      "auxiliary_loss_mlp": 0.0166616,
      "balance_loss_clip": 1.64215469,
      "balance_loss_mlp": 1.35850453,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 7.5786620770541475,
      "language_loss": 1.29310215,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.34022284,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.7906320095062256
    },
    {
      "auxiliary_loss_clip": 0.03041356,
      "auxiliary_loss_mlp": 0.01618372,
      "balance_loss_clip": 1.63828266,
      "balance_loss_mlp": 1.30594826,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.326214212927853,
      "language_loss": 1.13830554,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18490279,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 2.9531311988830566
    },
    {
      "auxiliary_loss_clip": 0.03017496,
      "auxiliary_loss_mlp": 0.01678042,
      "balance_loss_clip": 1.6332612,
      "balance_loss_mlp": 1.36161327,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.557942957666397,
      "language_loss": 1.12926197,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17621732,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.835988998413086
    },
    {
      "auxiliary_loss_clip": 0.03030541,
      "auxiliary_loss_mlp": 0.01453168,
      "balance_loss_clip": 1.77129507,
      "balance_loss_mlp": 1.26395857,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.7837975550139524,
      "language_loss": 0.64625597,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69109297,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 5.051850080490112
    },
    {
      "auxiliary_loss_clip": 0.03004118,
      "auxiliary_loss_mlp": 0.01637578,
      "balance_loss_clip": 1.63734543,
      "balance_loss_mlp": 1.32667983,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 7.868323652146635,
      "language_loss": 1.16303205,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20944905,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.766484022140503
    },
    {
      "auxiliary_loss_clip": 0.02996009,
      "auxiliary_loss_mlp": 0.01406174,
      "balance_loss_clip": 1.76380563,
      "balance_loss_mlp": 1.21925342,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.5115904515549827,
      "language_loss": 0.72122085,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.7652427,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.1702144145965576
    },
    {
      "auxiliary_loss_clip": 0.02973084,
      "auxiliary_loss_mlp": 0.01584428,
      "balance_loss_clip": 1.63080525,
      "balance_loss_mlp": 1.28306699,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 4.956989952636513,
      "language_loss": 1.01893091,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06450605,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.828373908996582
    },
    {
      "auxiliary_loss_clip": 0.02948643,
      "auxiliary_loss_mlp": 0.01368625,
      "balance_loss_clip": 1.75248146,
      "balance_loss_mlp": 1.18551946,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.7162270131858315,
      "language_loss": 0.76570845,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80888116,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 2.9903085231781006
    },
    {
      "auxiliary_loss_clip": 0.02925279,
      "auxiliary_loss_mlp": 0.01546167,
      "balance_loss_clip": 1.6208961,
      "balance_loss_mlp": 1.24575973,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 5.193287617285109,
      "language_loss": 1.02580822,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.07052255,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.7179410457611084
    },
    {
      "auxiliary_loss_clip": 0.02912053,
      "auxiliary_loss_mlp": 0.01528693,
      "balance_loss_clip": 1.61644244,
      "balance_loss_mlp": 1.23000264,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 22.66888734192237,
      "language_loss": 0.89860743,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.94301486,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.938603639602661
    },
    {
      "auxiliary_loss_clip": 0.02871706,
      "auxiliary_loss_mlp": 0.0151481,
      "balance_loss_clip": 1.60719228,
      "balance_loss_mlp": 1.21802628,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 6.293960327549222,
      "language_loss": 1.02962041,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.07348549,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.728699207305908
    },
    {
      "auxiliary_loss_clip": 0.0284295,
      "auxiliary_loss_mlp": 0.01490214,
      "balance_loss_clip": 1.59381688,
      "balance_loss_mlp": 1.19629204,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.4488838187805975,
      "language_loss": 1.03285837,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.07618999,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.7990524768829346
    },
    {
      "auxiliary_loss_clip": 0.02853261,
      "auxiliary_loss_mlp": 0.01489966,
      "balance_loss_clip": 1.60177565,
      "balance_loss_mlp": 1.20405483,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.0737488199523737,
      "language_loss": 0.92872643,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.97215873,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.845242500305176
    },
    {
      "auxiliary_loss_clip": 0.02809468,
      "auxiliary_loss_mlp": 0.01458336,
      "balance_loss_clip": 1.58923841,
      "balance_loss_mlp": 1.18081665,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.245426070029906,
      "language_loss": 0.93205351,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97473156,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.7795939445495605
    },
    {
      "auxiliary_loss_clip": 0.02818392,
      "auxiliary_loss_mlp": 0.014401,
      "balance_loss_clip": 1.59064388,
      "balance_loss_mlp": 1.16658664,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.685616016207989,
      "language_loss": 1.01149154,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05407643,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.8191983699798584
    },
    {
      "auxiliary_loss_clip": 0.02778492,
      "auxiliary_loss_mlp": 0.01444948,
      "balance_loss_clip": 1.58392644,
      "balance_loss_mlp": 1.15732014,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.738133733182587,
      "language_loss": 0.90894771,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.95118213,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.744889974594116
    },
    {
      "auxiliary_loss_clip": 0.02761996,
      "auxiliary_loss_mlp": 0.01418266,
      "balance_loss_clip": 1.57651651,
      "balance_loss_mlp": 1.14837623,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.26467687052534,
      "language_loss": 1.00204253,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.04384518,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.798753499984741
    },
    {
      "auxiliary_loss_clip": 0.02736204,
      "auxiliary_loss_mlp": 0.01418646,
      "balance_loss_clip": 1.56915736,
      "balance_loss_mlp": 1.14360666,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.1881393873655663,
      "language_loss": 0.9961257,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03767419,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.7880959510803223
    },
    {
      "auxiliary_loss_clip": 0.02750395,
      "auxiliary_loss_mlp": 0.0140321,
      "balance_loss_clip": 1.58317518,
      "balance_loss_mlp": 1.12149429,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 4.714566492024367,
      "language_loss": 1.00298572,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04452181,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.7917561531066895
    },
    {
      "auxiliary_loss_clip": 0.02730244,
      "auxiliary_loss_mlp": 0.01381907,
      "balance_loss_clip": 1.57163382,
      "balance_loss_mlp": 1.10782123,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.608317645385437,
      "language_loss": 0.99176121,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.03288269,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.7712419033050537
    },
    {
      "auxiliary_loss_clip": 0.02595321,
      "auxiliary_loss_mlp": 0.01366246,
      "balance_loss_clip": 1.53053629,
      "balance_loss_mlp": 1.09654737,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.6978274368329456,
      "language_loss": 1.02635419,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06596994,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.7650187015533447
    },
    {
      "auxiliary_loss_clip": 0.02546094,
      "auxiliary_loss_mlp": 0.01352393,
      "balance_loss_clip": 1.52295661,
      "balance_loss_mlp": 1.10539103,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 3.3596105253390807,
      "language_loss": 0.88184977,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.92083466,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.802351713180542
    },
    {
      "auxiliary_loss_clip": 0.02521737,
      "auxiliary_loss_mlp": 0.01329996,
      "balance_loss_clip": 1.51244569,
      "balance_loss_mlp": 1.07574677,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.116322424486873,
      "language_loss": 1.03525233,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07376969,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.8074567317962646
    },
    {
      "auxiliary_loss_clip": 0.02482584,
      "auxiliary_loss_mlp": 0.01358596,
      "balance_loss_clip": 1.50080645,
      "balance_loss_mlp": 1.10396504,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 2.008366831927587,
      "language_loss": 0.95128155,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98969328,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.7698676586151123
    },
    {
      "auxiliary_loss_clip": 0.02322323,
      "auxiliary_loss_mlp": 0.0139966,
      "balance_loss_clip": 1.58721638,
      "balance_loss_mlp": 1.25470185,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.3917030674864945,
      "language_loss": 0.60980618,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.647026,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.203026533126831
    },
    {
      "auxiliary_loss_clip": 0.0243217,
      "auxiliary_loss_mlp": 0.01364291,
      "balance_loss_clip": 1.49094439,
      "balance_loss_mlp": 1.11137676,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.164991929897605,
      "language_loss": 0.98987329,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02783799,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.7775299549102783
    },
    {
      "auxiliary_loss_clip": 0.02407184,
      "auxiliary_loss_mlp": 0.01330391,
      "balance_loss_clip": 1.4830209,
      "balance_loss_mlp": 1.093117,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.7854980014439175,
      "language_loss": 0.94180882,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97918463,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.7523300647735596
    },
    {
      "auxiliary_loss_clip": 0.02375688,
      "auxiliary_loss_mlp": 0.01336275,
      "balance_loss_clip": 1.48193109,
      "balance_loss_mlp": 1.09690285,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.614622108644164,
      "language_loss": 0.92619073,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96331036,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 2.810645818710327
    },
    {
      "auxiliary_loss_clip": 0.02367374,
      "auxiliary_loss_mlp": 0.01307282,
      "balance_loss_clip": 1.47209477,
      "balance_loss_mlp": 1.07897186,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.286629488815697,
      "language_loss": 0.93242437,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96917093,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 2.7651216983795166
    },
    {
      "auxiliary_loss_clip": 0.02333181,
      "auxiliary_loss_mlp": 0.01330038,
      "balance_loss_clip": 1.46068931,
      "balance_loss_mlp": 1.09676886,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 3.391209822133345,
      "language_loss": 0.99625254,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.03288484,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 3.731782913208008
    },
    {
      "auxiliary_loss_clip": 0.02306871,
      "auxiliary_loss_mlp": 0.01317557,
      "balance_loss_clip": 1.45267856,
      "balance_loss_mlp": 1.09363389,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 3.314984846096099,
      "language_loss": 0.92869359,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96493793,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 3.725327491760254
    },
    {
      "auxiliary_loss_clip": 0.02302326,
      "auxiliary_loss_mlp": 0.01314043,
      "balance_loss_clip": 1.45130944,
      "balance_loss_mlp": 1.09374416,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.3855586115626934,
      "language_loss": 0.83224905,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86841267,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.735771656036377
    },
    {
      "auxiliary_loss_clip": 0.02250139,
      "auxiliary_loss_mlp": 0.01334774,
      "balance_loss_clip": 1.43593621,
      "balance_loss_mlp": 1.11123276,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 23.309153869418044,
      "language_loss": 0.79304636,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82889557,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.7580983638763428
    },
    {
      "auxiliary_loss_clip": 0.02237372,
      "auxiliary_loss_mlp": 0.01311841,
      "balance_loss_clip": 1.43013144,
      "balance_loss_mlp": 1.0938307,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 1.977508921787129,
      "language_loss": 1.04310441,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07859659,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.8093459606170654
    },
    {
      "auxiliary_loss_clip": 0.02226355,
      "auxiliary_loss_mlp": 0.0127948,
      "balance_loss_clip": 1.42991853,
      "balance_loss_mlp": 1.06623793,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.1164711507489398,
      "language_loss": 1.07998824,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11504662,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.7489101886749268
    },
    {
      "auxiliary_loss_clip": 0.02190741,
      "auxiliary_loss_mlp": 0.01288463,
      "balance_loss_clip": 1.41754222,
      "balance_loss_mlp": 1.07932198,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.582539623671459,
      "language_loss": 1.08986306,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12465513,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.768611431121826
    },
    {
      "auxiliary_loss_clip": 0.02164875,
      "auxiliary_loss_mlp": 0.01284313,
      "balance_loss_clip": 1.41716421,
      "balance_loss_mlp": 1.07393229,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 2.9689744494371446,
      "language_loss": 1.02182293,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05631483,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.7494699954986572
    },
    {
      "auxiliary_loss_clip": 0.02155902,
      "auxiliary_loss_mlp": 0.01295158,
      "balance_loss_clip": 1.41105092,
      "balance_loss_mlp": 1.09383738,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 2.9889148814922644,
      "language_loss": 0.90513557,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93964618,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.7773077487945557
    },
    {
      "auxiliary_loss_clip": 0.02150929,
      "auxiliary_loss_mlp": 0.01258099,
      "balance_loss_clip": 1.40764952,
      "balance_loss_mlp": 1.06269097,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 2.449163469651116,
      "language_loss": 0.92609608,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.9601863,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.732020139694214
    },
    {
      "auxiliary_loss_clip": 0.02132962,
      "auxiliary_loss_mlp": 0.01297469,
      "balance_loss_clip": 1.40845942,
      "balance_loss_mlp": 1.09796,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 10.833833705908543,
      "language_loss": 0.978585,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.01288927,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.7678189277648926
    },
    {
      "auxiliary_loss_clip": 0.0211633,
      "auxiliary_loss_mlp": 0.01265646,
      "balance_loss_clip": 1.39879084,
      "balance_loss_mlp": 1.078058,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.04343359818791,
      "language_loss": 0.89086783,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.9246875,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.7815988063812256
    },
    {
      "auxiliary_loss_clip": 0.02113809,
      "auxiliary_loss_mlp": 0.01280719,
      "balance_loss_clip": 1.40224206,
      "balance_loss_mlp": 1.09027052,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 3.38362059960225,
      "language_loss": 0.92717862,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.96112388,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.787706136703491
    },
    {
      "auxiliary_loss_clip": 0.02082962,
      "auxiliary_loss_mlp": 0.01301368,
      "balance_loss_clip": 1.39041257,
      "balance_loss_mlp": 1.10128748,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.983562460100215,
      "language_loss": 0.84677619,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88061953,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.7874155044555664
    },
    {
      "auxiliary_loss_clip": 0.02083816,
      "auxiliary_loss_mlp": 0.01256511,
      "balance_loss_clip": 1.39221358,
      "balance_loss_mlp": 1.07607615,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 1.9185944341333192,
      "language_loss": 0.91829759,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95170093,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.7796990871429443
    },
    {
      "auxiliary_loss_clip": 0.02064299,
      "auxiliary_loss_mlp": 0.01257157,
      "balance_loss_clip": 1.3861326,
      "balance_loss_mlp": 1.07500482,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.5795990915026725,
      "language_loss": 0.86240828,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89562285,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.8482019901275635
    },
    {
      "auxiliary_loss_clip": 0.02035068,
      "auxiliary_loss_mlp": 0.01237735,
      "balance_loss_clip": 1.3767221,
      "balance_loss_mlp": 1.06645501,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 1.8766334015780493,
      "language_loss": 1.01981592,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.052544,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.815507173538208
    },
    {
      "auxiliary_loss_clip": 0.02050471,
      "auxiliary_loss_mlp": 0.01259909,
      "balance_loss_clip": 1.37825072,
      "balance_loss_mlp": 1.08748496,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 2.229407759359226,
      "language_loss": 0.93603319,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96913701,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.8137729167938232
    },
    {
      "auxiliary_loss_clip": 0.0201911,
      "auxiliary_loss_mlp": 0.01245275,
      "balance_loss_clip": 1.37091708,
      "balance_loss_mlp": 1.07771397,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.2335288499984394,
      "language_loss": 0.97673583,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00937974,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.692969799041748
    },
    {
      "auxiliary_loss_clip": 0.01995283,
      "auxiliary_loss_mlp": 0.01241532,
      "balance_loss_clip": 1.3681525,
      "balance_loss_mlp": 1.07854867,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.5119726809599046,
      "language_loss": 0.93803644,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97040462,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.762362480163574
    },
    {
      "auxiliary_loss_clip": 0.01981269,
      "auxiliary_loss_mlp": 0.01255103,
      "balance_loss_clip": 1.36431623,
      "balance_loss_mlp": 1.0915482,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 4.531889711918123,
      "language_loss": 0.72302353,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75538731,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.7043561935424805
    },
    {
      "auxiliary_loss_clip": 0.01983,
      "auxiliary_loss_mlp": 0.01256037,
      "balance_loss_clip": 1.36239958,
      "balance_loss_mlp": 1.08828604,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 2.6569921957737974,
      "language_loss": 0.88157368,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91396403,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.7512130737304688
    },
    {
      "auxiliary_loss_clip": 0.01957829,
      "auxiliary_loss_mlp": 0.01228119,
      "balance_loss_clip": 1.35374153,
      "balance_loss_mlp": 1.07667518,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.666898894486535,
      "language_loss": 0.96433914,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99619859,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.798231601715088
    },
    {
      "auxiliary_loss_clip": 0.01962679,
      "auxiliary_loss_mlp": 0.01231443,
      "balance_loss_clip": 1.3555057,
      "balance_loss_mlp": 1.07265592,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 1.9298201176111924,
      "language_loss": 0.88196731,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91390848,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.7382514476776123
    },
    {
      "auxiliary_loss_clip": 0.01942951,
      "auxiliary_loss_mlp": 0.0125731,
      "balance_loss_clip": 1.34462643,
      "balance_loss_mlp": 1.09108448,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.343885306554678,
      "language_loss": 0.87653768,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90854025,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 2.763967275619507
    },
    {
      "auxiliary_loss_clip": 0.01922659,
      "auxiliary_loss_mlp": 0.01217703,
      "balance_loss_clip": 1.33828831,
      "balance_loss_mlp": 1.07999301,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.0968227762011282,
      "language_loss": 0.88646424,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91786784,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 2.7323243618011475
    },
    {
      "auxiliary_loss_clip": 0.01916553,
      "auxiliary_loss_mlp": 0.01217674,
      "balance_loss_clip": 1.3369174,
      "balance_loss_mlp": 1.07376409,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.6283864101064585,
      "language_loss": 1.04869485,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.08003712,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.6945388317108154
    },
    {
      "auxiliary_loss_clip": 0.01898749,
      "auxiliary_loss_mlp": 0.01220646,
      "balance_loss_clip": 1.33036625,
      "balance_loss_mlp": 1.08036041,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.9982707731762006,
      "language_loss": 0.95438772,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98558164,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 4.590542554855347
    },
    {
      "auxiliary_loss_clip": 0.01753018,
      "auxiliary_loss_mlp": 0.01223789,
      "balance_loss_clip": 1.35544157,
      "balance_loss_mlp": 1.15436113,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.3005417665230021,
      "language_loss": 0.62118924,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.65095735,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 4.282896041870117
    },
    {
      "auxiliary_loss_clip": 0.01869418,
      "auxiliary_loss_mlp": 0.01211422,
      "balance_loss_clip": 1.3118068,
      "balance_loss_mlp": 1.07361627,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.177383320688634,
      "language_loss": 0.90474176,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93555021,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.785377264022827
    },
    {
      "auxiliary_loss_clip": 0.01718998,
      "auxiliary_loss_mlp": 0.0118351,
      "balance_loss_clip": 1.33499932,
      "balance_loss_mlp": 1.11866021,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.166799713600779,
      "language_loss": 0.5856353,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61466038,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.1999244689941406
    },
    {
      "auxiliary_loss_clip": 0.01858922,
      "auxiliary_loss_mlp": 0.01210212,
      "balance_loss_clip": 1.30900121,
      "balance_loss_mlp": 1.07898617,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.328647281922714,
      "language_loss": 0.98947358,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.02016485,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.8093578815460205
    },
    {
      "auxiliary_loss_clip": 0.01829755,
      "auxiliary_loss_mlp": 0.01199867,
      "balance_loss_clip": 1.29966187,
      "balance_loss_mlp": 1.0657804,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 4.979081430159531,
      "language_loss": 0.96618462,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99648082,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.735330820083618
    },
    {
      "auxiliary_loss_clip": 0.01814588,
      "auxiliary_loss_mlp": 0.01209057,
      "balance_loss_clip": 1.29189873,
      "balance_loss_mlp": 1.08307612,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 2.718044155792231,
      "language_loss": 0.80164671,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83188313,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.7860264778137207
    },
    {
      "auxiliary_loss_clip": 0.01793737,
      "auxiliary_loss_mlp": 0.01179664,
      "balance_loss_clip": 1.2845881,
      "balance_loss_mlp": 1.06398344,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.0006588236814826,
      "language_loss": 0.95645761,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98619157,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.925122022628784
    },
    {
      "auxiliary_loss_clip": 0.01791494,
      "auxiliary_loss_mlp": 0.01210157,
      "balance_loss_clip": 1.28563726,
      "balance_loss_mlp": 1.0873239,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 2.107231104863039,
      "language_loss": 0.79676032,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82677686,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.789067268371582
    },
    {
      "auxiliary_loss_clip": 0.0178176,
      "auxiliary_loss_mlp": 0.01193952,
      "balance_loss_clip": 1.27956414,
      "balance_loss_mlp": 1.07321692,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.747223904525475,
      "language_loss": 0.91689658,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94665372,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.752852439880371
    },
    {
      "auxiliary_loss_clip": 0.0177732,
      "auxiliary_loss_mlp": 0.01188281,
      "balance_loss_clip": 1.27649212,
      "balance_loss_mlp": 1.06830859,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 3.62207735639936,
      "language_loss": 0.74985009,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77950609,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.831871509552002
    },
    {
      "auxiliary_loss_clip": 0.01757183,
      "auxiliary_loss_mlp": 0.01179658,
      "balance_loss_clip": 1.27056742,
      "balance_loss_mlp": 1.06989014,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 5.267081396000725,
      "language_loss": 0.76007986,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.78944826,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.6933987140655518
    },
    {
      "auxiliary_loss_clip": 0.01746328,
      "auxiliary_loss_mlp": 0.01174525,
      "balance_loss_clip": 1.26579428,
      "balance_loss_mlp": 1.06280208,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 2.992231167061448,
      "language_loss": 0.87279755,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.90200615,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.7544960975646973
    },
    {
      "auxiliary_loss_clip": 0.0172579,
      "auxiliary_loss_mlp": 0.0118876,
      "balance_loss_clip": 1.25895047,
      "balance_loss_mlp": 1.07875371,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.4378176868080024,
      "language_loss": 0.8936004,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92274594,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.761899709701538
    },
    {
      "auxiliary_loss_clip": 0.01735378,
      "auxiliary_loss_mlp": 0.01186964,
      "balance_loss_clip": 1.26342249,
      "balance_loss_mlp": 1.07748258,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.251383410462456,
      "language_loss": 0.83435601,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86357945,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.7250611782073975
    },
    {
      "auxiliary_loss_clip": 0.01722802,
      "auxiliary_loss_mlp": 0.01176101,
      "balance_loss_clip": 1.25636911,
      "balance_loss_mlp": 1.06967127,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.5721591726057684,
      "language_loss": 0.95242196,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.98141098,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.717829465866089
    },
    {
      "auxiliary_loss_clip": 0.01720745,
      "auxiliary_loss_mlp": 0.01181897,
      "balance_loss_clip": 1.25805044,
      "balance_loss_mlp": 1.07951999,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.802882052878897,
      "language_loss": 0.89328706,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92231345,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.742253541946411
    },
    {
      "auxiliary_loss_clip": 0.01719379,
      "auxiliary_loss_mlp": 0.01184069,
      "balance_loss_clip": 1.2586534,
      "balance_loss_mlp": 1.08293235,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.1830337582958346,
      "language_loss": 0.93303239,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.96206683,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.8064610958099365
    },
    {
      "auxiliary_loss_clip": 0.01703418,
      "auxiliary_loss_mlp": 0.0117541,
      "balance_loss_clip": 1.25011384,
      "balance_loss_mlp": 1.08023369,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.1690057839926165,
      "language_loss": 0.96494353,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99373186,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.7376246452331543
    },
    {
      "auxiliary_loss_clip": 0.01681868,
      "auxiliary_loss_mlp": 0.01169618,
      "balance_loss_clip": 1.24167144,
      "balance_loss_mlp": 1.0804019,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 1.9510716959253074,
      "language_loss": 0.81393772,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84245253,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.8326058387756348
    },
    {
      "auxiliary_loss_clip": 0.01689032,
      "auxiliary_loss_mlp": 0.01158295,
      "balance_loss_clip": 1.24723327,
      "balance_loss_mlp": 1.06473935,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 2.1706096515977973,
      "language_loss": 0.94106424,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.9695375,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.710801601409912
    },
    {
      "auxiliary_loss_clip": 0.01675073,
      "auxiliary_loss_mlp": 0.01157932,
      "balance_loss_clip": 1.23361003,
      "balance_loss_mlp": 1.06957388,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 3.2608328727107785,
      "language_loss": 0.75271249,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.78104252,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.7228455543518066
    },
    {
      "auxiliary_loss_clip": 0.01683233,
      "auxiliary_loss_mlp": 0.01190832,
      "balance_loss_clip": 1.24348271,
      "balance_loss_mlp": 1.09923196,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.1518849166427043,
      "language_loss": 0.84885365,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87759435,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.7231569290161133
    },
    {
      "auxiliary_loss_clip": 0.01665375,
      "auxiliary_loss_mlp": 0.01168478,
      "balance_loss_clip": 1.23648858,
      "balance_loss_mlp": 1.08269513,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.676724513729159,
      "language_loss": 0.92021191,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94855046,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 2.80464243888855
    },
    {
      "auxiliary_loss_clip": 0.01653591,
      "auxiliary_loss_mlp": 0.01150477,
      "balance_loss_clip": 1.22836673,
      "balance_loss_mlp": 1.0717032,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.0866827642936316,
      "language_loss": 0.96190202,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98994267,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 2.68426775932312
    },
    {
      "auxiliary_loss_clip": 0.01650803,
      "auxiliary_loss_mlp": 0.01179505,
      "balance_loss_clip": 1.22757125,
      "balance_loss_mlp": 1.08561635,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 2.495533027429512,
      "language_loss": 0.85665321,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88495624,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.7060933113098145
    },
    {
      "auxiliary_loss_clip": 0.01646414,
      "auxiliary_loss_mlp": 0.01157415,
      "balance_loss_clip": 1.22522414,
      "balance_loss_mlp": 1.0779736,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.165279990114633,
      "language_loss": 0.8878305,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91586882,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.6954009532928467
    },
    {
      "auxiliary_loss_clip": 0.01646638,
      "auxiliary_loss_mlp": 0.01143243,
      "balance_loss_clip": 1.23007619,
      "balance_loss_mlp": 1.0619421,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 1.740222449960583,
      "language_loss": 0.80184293,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82974172,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 4.7059173583984375
    },
    {
      "auxiliary_loss_clip": 0.01626106,
      "auxiliary_loss_mlp": 0.01142295,
      "balance_loss_clip": 1.21771574,
      "balance_loss_mlp": 1.06552374,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 8.692964296973088,
      "language_loss": 0.92267519,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95035928,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 3.7725651264190674
    },
    {
      "auxiliary_loss_clip": 0.01630536,
      "auxiliary_loss_mlp": 0.01147011,
      "balance_loss_clip": 1.21399045,
      "balance_loss_mlp": 1.07076526,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.660651452302514,
      "language_loss": 0.86868083,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89645636,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.723741054534912
    },
    {
      "auxiliary_loss_clip": 0.0162069,
      "auxiliary_loss_mlp": 0.01138543,
      "balance_loss_clip": 1.21725559,
      "balance_loss_mlp": 1.07083178,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.031282339782685,
      "language_loss": 1.01719618,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04478848,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.69109845161438
    },
    {
      "auxiliary_loss_clip": 0.01618002,
      "auxiliary_loss_mlp": 0.01137213,
      "balance_loss_clip": 1.21361399,
      "balance_loss_mlp": 1.06292152,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.181764444589878,
      "language_loss": 0.97679728,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00434947,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.80876088142395
    },
    {
      "auxiliary_loss_clip": 0.01609573,
      "auxiliary_loss_mlp": 0.01124988,
      "balance_loss_clip": 1.21018815,
      "balance_loss_mlp": 1.0597564,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.2333159671824308,
      "language_loss": 0.93745786,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96480346,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.661630630493164
    },
    {
      "auxiliary_loss_clip": 0.01612168,
      "auxiliary_loss_mlp": 0.01133708,
      "balance_loss_clip": 1.20975661,
      "balance_loss_mlp": 1.06671286,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 2.590385742161331,
      "language_loss": 0.81117243,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83863127,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.713487386703491
    },
    {
      "auxiliary_loss_clip": 0.01605354,
      "auxiliary_loss_mlp": 0.01149923,
      "balance_loss_clip": 1.20347357,
      "balance_loss_mlp": 1.08340394,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.530493434940448,
      "language_loss": 0.87573314,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90328592,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.6449739933013916
    },
    {
      "auxiliary_loss_clip": 0.01593706,
      "auxiliary_loss_mlp": 0.01136416,
      "balance_loss_clip": 1.20098114,
      "balance_loss_mlp": 1.06887209,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.1257646389750913,
      "language_loss": 0.88200629,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90930754,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.768324613571167
    },
    {
      "auxiliary_loss_clip": 0.01585191,
      "auxiliary_loss_mlp": 0.01136189,
      "balance_loss_clip": 1.19637847,
      "balance_loss_mlp": 1.06566513,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 2.0729959826335715,
      "language_loss": 0.8557409,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.8829546,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.818066120147705
    },
    {
      "auxiliary_loss_clip": 0.01588762,
      "auxiliary_loss_mlp": 0.01176536,
      "balance_loss_clip": 1.19927394,
      "balance_loss_mlp": 1.10868168,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 2.871253785953426,
      "language_loss": 0.90116918,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92882216,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.704796075820923
    },
    {
      "auxiliary_loss_clip": 0.01579407,
      "auxiliary_loss_mlp": 0.01138483,
      "balance_loss_clip": 1.19024765,
      "balance_loss_mlp": 1.07129669,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 1.9278255351972964,
      "language_loss": 0.88621694,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91339582,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.7234060764312744
    },
    {
      "auxiliary_loss_clip": 0.01574585,
      "auxiliary_loss_mlp": 0.01128419,
      "balance_loss_clip": 1.18907428,
      "balance_loss_mlp": 1.06409431,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.238181736820889,
      "language_loss": 0.96235865,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98938859,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.7696850299835205
    },
    {
      "auxiliary_loss_clip": 0.0156392,
      "auxiliary_loss_mlp": 0.01125695,
      "balance_loss_clip": 1.18485713,
      "balance_loss_mlp": 1.06671023,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 2.1560987297492975,
      "language_loss": 0.86794448,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.8948406,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.727442741394043
    },
    {
      "auxiliary_loss_clip": 0.0155868,
      "auxiliary_loss_mlp": 0.01107997,
      "balance_loss_clip": 1.18197656,
      "balance_loss_mlp": 1.04996634,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.632131214212815,
      "language_loss": 0.95261455,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.97928131,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.65803599357605
    },
    {
      "auxiliary_loss_clip": 0.01559317,
      "auxiliary_loss_mlp": 0.01126871,
      "balance_loss_clip": 1.18487823,
      "balance_loss_mlp": 1.06659853,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 2.123787144559695,
      "language_loss": 0.81196761,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.83882952,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.730092763900757
    },
    {
      "auxiliary_loss_clip": 0.01562539,
      "auxiliary_loss_mlp": 0.01125747,
      "balance_loss_clip": 1.1836158,
      "balance_loss_mlp": 1.06776381,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 1.762805318570285,
      "language_loss": 0.75978386,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78666675,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.701456069946289
    },
    {
      "auxiliary_loss_clip": 0.01550302,
      "auxiliary_loss_mlp": 0.01145233,
      "balance_loss_clip": 1.17949271,
      "balance_loss_mlp": 1.0872494,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.50358446644987,
      "language_loss": 0.95044601,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97740132,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.733604907989502
    },
    {
      "auxiliary_loss_clip": 0.01553212,
      "auxiliary_loss_mlp": 0.01139942,
      "balance_loss_clip": 1.17959976,
      "balance_loss_mlp": 1.08026612,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 1.89993948959094,
      "language_loss": 0.88572758,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91265911,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.71427583694458
    },
    {
      "auxiliary_loss_clip": 0.01538706,
      "auxiliary_loss_mlp": 0.01128228,
      "balance_loss_clip": 1.17495918,
      "balance_loss_mlp": 1.07339191,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 1.6480570901306548,
      "language_loss": 0.77567208,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80234152,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.8351428508758545
    },
    {
      "auxiliary_loss_clip": 0.01534961,
      "auxiliary_loss_mlp": 0.01091467,
      "balance_loss_clip": 1.16938567,
      "balance_loss_mlp": 1.04035056,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.0151060859341663,
      "language_loss": 0.90896547,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93522978,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.83171010017395
    },
    {
      "auxiliary_loss_clip": 0.01530879,
      "auxiliary_loss_mlp": 0.01106792,
      "balance_loss_clip": 1.16982365,
      "balance_loss_mlp": 1.05798745,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.1115856665680086,
      "language_loss": 0.86176407,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.8881408,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.693544387817383
    },
    {
      "auxiliary_loss_clip": 0.01532449,
      "auxiliary_loss_mlp": 0.01111775,
      "balance_loss_clip": 1.17066908,
      "balance_loss_mlp": 1.06063378,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.7216706157159627,
      "language_loss": 0.89180529,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.91824758,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 2.7406249046325684
    },
    {
      "auxiliary_loss_clip": 0.01519612,
      "auxiliary_loss_mlp": 0.0112149,
      "balance_loss_clip": 1.16331935,
      "balance_loss_mlp": 1.07237554,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.873578454231734,
      "language_loss": 0.92401195,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.95042294,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 2.7279345989227295
    },
    {
      "auxiliary_loss_clip": 0.01538669,
      "auxiliary_loss_mlp": 0.01127136,
      "balance_loss_clip": 1.17217767,
      "balance_loss_mlp": 1.07711625,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.6561431373666524,
      "language_loss": 0.85255051,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.87920856,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 2.696613311767578
    },
    {
      "auxiliary_loss_clip": 0.01513505,
      "auxiliary_loss_mlp": 0.0110602,
      "balance_loss_clip": 1.16414809,
      "balance_loss_mlp": 1.06479692,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 1.8368742885819591,
      "language_loss": 0.83023602,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85643125,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.6497819423675537
    },
    {
      "auxiliary_loss_clip": 0.01511881,
      "auxiliary_loss_mlp": 0.01116001,
      "balance_loss_clip": 1.16216695,
      "balance_loss_mlp": 1.07170248,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.1149151695392936,
      "language_loss": 0.8855201,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91179895,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.6918535232543945
    },
    {
      "auxiliary_loss_clip": 0.01515049,
      "auxiliary_loss_mlp": 0.01109849,
      "balance_loss_clip": 1.16170585,
      "balance_loss_mlp": 1.06500268,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 1.9573995361694998,
      "language_loss": 0.89511579,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92136478,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 5.445294141769409
    },
    {
      "auxiliary_loss_clip": 0.01505972,
      "auxiliary_loss_mlp": 0.01105081,
      "balance_loss_clip": 1.15691829,
      "balance_loss_mlp": 1.059376,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 1.961550922088161,
      "language_loss": 0.8408947,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86700523,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 3.636497974395752
    },
    {
      "auxiliary_loss_clip": 0.0150149,
      "auxiliary_loss_mlp": 0.01105957,
      "balance_loss_clip": 1.15377665,
      "balance_loss_mlp": 1.06442451,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.515651904866598,
      "language_loss": 0.87069464,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89676905,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.6189591884613037
    },
    {
      "auxiliary_loss_clip": 0.01500161,
      "auxiliary_loss_mlp": 0.01113203,
      "balance_loss_clip": 1.1553967,
      "balance_loss_mlp": 1.07300591,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 2.0451193549464346,
      "language_loss": 0.96394598,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.9900797,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.6986865997314453
    },
    {
      "auxiliary_loss_clip": 0.01500794,
      "auxiliary_loss_mlp": 0.0110829,
      "balance_loss_clip": 1.15414047,
      "balance_loss_mlp": 1.06427753,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.8304376015016706,
      "language_loss": 0.82674122,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85283208,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.839111566543579
    },
    {
      "auxiliary_loss_clip": 0.01498262,
      "auxiliary_loss_mlp": 0.01119548,
      "balance_loss_clip": 1.15211749,
      "balance_loss_mlp": 1.07465386,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.1675368688267955,
      "language_loss": 0.78476572,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.81094378,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.841923236846924
    },
    {
      "auxiliary_loss_clip": 0.01494094,
      "auxiliary_loss_mlp": 0.01088525,
      "balance_loss_clip": 1.15283942,
      "balance_loss_mlp": 1.05116475,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.8894902473602804,
      "language_loss": 0.84112978,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.866956,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.6371238231658936
    },
    {
      "auxiliary_loss_clip": 0.01496292,
      "auxiliary_loss_mlp": 0.01113386,
      "balance_loss_clip": 1.15502191,
      "balance_loss_mlp": 1.0732367,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.8491519678638664,
      "language_loss": 0.87622505,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90232182,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.8094050884246826
    },
    {
      "auxiliary_loss_clip": 0.01459008,
      "auxiliary_loss_mlp": 0.01062512,
      "balance_loss_clip": 1.17543507,
      "balance_loss_mlp": 1.04820716,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.04764098252005,
      "language_loss": 0.71533215,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.7405473,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.2729580402374268
    },
    {
      "auxiliary_loss_clip": 0.01481828,
      "auxiliary_loss_mlp": 0.01120884,
      "balance_loss_clip": 1.14767504,
      "balance_loss_mlp": 1.08330917,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 3.5714132709935407,
      "language_loss": 0.82397771,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.85000479,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.69273042678833
    },
    {
      "auxiliary_loss_clip": 0.0147947,
      "auxiliary_loss_mlp": 0.01112361,
      "balance_loss_clip": 1.14617872,
      "balance_loss_mlp": 1.0747627,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 8.480037360449188,
      "language_loss": 0.90295112,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92886943,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.72808575630188
    },
    {
      "auxiliary_loss_clip": 0.01478736,
      "auxiliary_loss_mlp": 0.01105158,
      "balance_loss_clip": 1.14653206,
      "balance_loss_mlp": 1.06670094,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.2723405680555193,
      "language_loss": 0.87139797,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89723694,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.6177892684936523
    },
    {
      "auxiliary_loss_clip": 0.01477611,
      "auxiliary_loss_mlp": 0.01100664,
      "balance_loss_clip": 1.14705014,
      "balance_loss_mlp": 1.06285036,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.0522808414994973,
      "language_loss": 0.76109636,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78687912,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.767521381378174
    },
    {
      "auxiliary_loss_clip": 0.014656,
      "auxiliary_loss_mlp": 0.0109919,
      "balance_loss_clip": 1.13992405,
      "balance_loss_mlp": 1.06211567,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 1.8296759633810094,
      "language_loss": 0.83244491,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85809278,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.6587913036346436
    },
    {
      "auxiliary_loss_clip": 0.0147412,
      "auxiliary_loss_mlp": 0.01099946,
      "balance_loss_clip": 1.14243364,
      "balance_loss_mlp": 1.06280017,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 1.9911819932097807,
      "language_loss": 0.77041364,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79615426,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.686795711517334
    },
    {
      "auxiliary_loss_clip": 0.01469472,
      "auxiliary_loss_mlp": 0.01090961,
      "balance_loss_clip": 1.1489892,
      "balance_loss_mlp": 1.05932283,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 3.1101496617082667,
      "language_loss": 0.83953768,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86514199,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.6461851596832275
    },
    {
      "auxiliary_loss_clip": 0.01471159,
      "auxiliary_loss_mlp": 0.01112497,
      "balance_loss_clip": 1.14420784,
      "balance_loss_mlp": 1.07411146,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 3.2007533729515765,
      "language_loss": 0.89913177,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92496836,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.64290189743042
    },
    {
      "auxiliary_loss_clip": 0.01476055,
      "auxiliary_loss_mlp": 0.01102314,
      "balance_loss_clip": 1.14604449,
      "balance_loss_mlp": 1.06605053,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.525983734056995,
      "language_loss": 0.90727139,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93305504,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.545984983444214
    },
    {
      "auxiliary_loss_clip": 0.01458295,
      "auxiliary_loss_mlp": 0.01104618,
      "balance_loss_clip": 1.13670588,
      "balance_loss_mlp": 1.07050037,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 1.9126127830983397,
      "language_loss": 0.87981719,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90544629,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.6115992069244385
    },
    {
      "auxiliary_loss_clip": 0.0146505,
      "auxiliary_loss_mlp": 0.01100854,
      "balance_loss_clip": 1.14048815,
      "balance_loss_mlp": 1.0662359,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 1.7690248893196379,
      "language_loss": 0.78423345,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.80989254,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.5733721256256104
    },
    {
      "auxiliary_loss_clip": 0.01456534,
      "auxiliary_loss_mlp": 0.01104693,
      "balance_loss_clip": 1.13813734,
      "balance_loss_mlp": 1.07224393,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.0924676759557457,
      "language_loss": 0.87240976,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.89802206,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.5568275451660156
    },
    {
      "auxiliary_loss_clip": 0.01453171,
      "auxiliary_loss_mlp": 0.01087538,
      "balance_loss_clip": 1.13440919,
      "balance_loss_mlp": 1.0564121,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 1.8389217653698877,
      "language_loss": 0.80640322,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.8318103,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.692556858062744
    },
    {
      "auxiliary_loss_clip": 0.01457865,
      "auxiliary_loss_mlp": 0.01089138,
      "balance_loss_clip": 1.1357224,
      "balance_loss_mlp": 1.05585527,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 2.3000401788365847,
      "language_loss": 0.97185981,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99732995,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 2.6630749702453613
    },
    {
      "auxiliary_loss_clip": 0.01407653,
      "auxiliary_loss_mlp": 0.01015233,
      "balance_loss_clip": 1.15434265,
      "balance_loss_mlp": 1.0031693,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0400607833169153,
      "language_loss": 0.58450389,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.6087327,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.2007577419281006
    },
    {
      "auxiliary_loss_clip": 0.0144818,
      "auxiliary_loss_mlp": 0.01087058,
      "balance_loss_clip": 1.13325417,
      "balance_loss_mlp": 1.05631435,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 3.493728562104307,
      "language_loss": 0.85442793,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.87978041,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.697659969329834
    },
    {
      "auxiliary_loss_clip": 0.01445633,
      "auxiliary_loss_mlp": 0.01099063,
      "balance_loss_clip": 1.13079,
      "balance_loss_mlp": 1.06873631,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.3616006335284423,
      "language_loss": 0.88640034,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91184729,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.703683853149414
    },
    {
      "auxiliary_loss_clip": 0.01440911,
      "auxiliary_loss_mlp": 0.01095122,
      "balance_loss_clip": 1.12952375,
      "balance_loss_mlp": 1.06572485,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 1.872428792181536,
      "language_loss": 0.8822484,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90760875,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 3.544456720352173
    },
    {
      "auxiliary_loss_clip": 0.01455089,
      "auxiliary_loss_mlp": 0.01099102,
      "balance_loss_clip": 1.13743377,
      "balance_loss_mlp": 1.06956196,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 2.5006175559263815,
      "language_loss": 0.86493683,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.89047873,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 4.636069059371948
    },
    {
      "auxiliary_loss_clip": 0.01447614,
      "auxiliary_loss_mlp": 0.0108766,
      "balance_loss_clip": 1.13441706,
      "balance_loss_mlp": 1.05950308,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.5317854658375665,
      "language_loss": 0.93804634,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96339905,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 3.5446135997772217
    },
    {
      "auxiliary_loss_clip": 0.01439596,
      "auxiliary_loss_mlp": 0.01088944,
      "balance_loss_clip": 1.13139105,
      "balance_loss_mlp": 1.0608108,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.6235318844281268,
      "language_loss": 0.85625875,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88154417,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.8105549812316895
    },
    {
      "auxiliary_loss_clip": 0.01441435,
      "auxiliary_loss_mlp": 0.01080649,
      "balance_loss_clip": 1.13119221,
      "balance_loss_mlp": 1.05096555,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 15.251962589374024,
      "language_loss": 0.8887428,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91396368,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.6447861194610596
    },
    {
      "auxiliary_loss_clip": 0.01444899,
      "auxiliary_loss_mlp": 0.01094685,
      "balance_loss_clip": 1.13133276,
      "balance_loss_mlp": 1.06384587,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.2235506235236264,
      "language_loss": 0.95911109,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98450691,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.705242872238159
    },
    {
      "auxiliary_loss_clip": 0.01431867,
      "auxiliary_loss_mlp": 0.01090042,
      "balance_loss_clip": 1.1273005,
      "balance_loss_mlp": 1.06392312,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 2.0366553740675686,
      "language_loss": 0.90631139,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93153048,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.6305992603302
    },
    {
      "auxiliary_loss_clip": 0.01435983,
      "auxiliary_loss_mlp": 0.01070916,
      "balance_loss_clip": 1.12564492,
      "balance_loss_mlp": 1.04284191,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 1.8871226565251273,
      "language_loss": 0.75898361,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78405255,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.7368271350860596
    },
    {
      "auxiliary_loss_clip": 0.01430425,
      "auxiliary_loss_mlp": 0.01080069,
      "balance_loss_clip": 1.12496114,
      "balance_loss_mlp": 1.05275798,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.731313691305322,
      "language_loss": 1.0409019,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06600666,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.622145891189575
    },
    {
      "auxiliary_loss_clip": 0.01427603,
      "auxiliary_loss_mlp": 0.01080316,
      "balance_loss_clip": 1.12679601,
      "balance_loss_mlp": 1.05293357,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.183176538736381,
      "language_loss": 0.8530513,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.87813056,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.6507339477539062
    },
    {
      "auxiliary_loss_clip": 0.01421258,
      "auxiliary_loss_mlp": 0.01066211,
      "balance_loss_clip": 1.12074447,
      "balance_loss_mlp": 1.04097486,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.4024841907217143,
      "language_loss": 0.88725197,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91212666,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.70408296585083
    },
    {
      "auxiliary_loss_clip": 0.0142881,
      "auxiliary_loss_mlp": 0.01077699,
      "balance_loss_clip": 1.12867248,
      "balance_loss_mlp": 1.05307007,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 1.8474949790028932,
      "language_loss": 0.86366975,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88873482,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.628230571746826
    },
    {
      "auxiliary_loss_clip": 0.01428543,
      "auxiliary_loss_mlp": 0.01081061,
      "balance_loss_clip": 1.12965417,
      "balance_loss_mlp": 1.05626512,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 2.726798063233743,
      "language_loss": 0.87678206,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.901878,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.6736960411071777
    },
    {
      "auxiliary_loss_clip": 0.01422417,
      "auxiliary_loss_mlp": 0.01082761,
      "balance_loss_clip": 1.12482333,
      "balance_loss_mlp": 1.05871689,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 3.188227268290155,
      "language_loss": 0.81269336,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.83774519,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.6389787197113037
    },
    {
      "auxiliary_loss_clip": 0.01422046,
      "auxiliary_loss_mlp": 0.01086273,
      "balance_loss_clip": 1.12568462,
      "balance_loss_mlp": 1.06250286,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.3606725487562525,
      "language_loss": 0.93961418,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96469736,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.6831021308898926
    },
    {
      "auxiliary_loss_clip": 0.01424933,
      "auxiliary_loss_mlp": 0.0106776,
      "balance_loss_clip": 1.1246196,
      "balance_loss_mlp": 1.04148674,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.185243033678872,
      "language_loss": 0.87153018,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.8964572,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.679617404937744
    },
    {
      "auxiliary_loss_clip": 0.01425191,
      "auxiliary_loss_mlp": 0.01080412,
      "balance_loss_clip": 1.12682009,
      "balance_loss_mlp": 1.05533004,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.3446358844323334,
      "language_loss": 0.73639798,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76145399,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.6329104900360107
    },
    {
      "auxiliary_loss_clip": 0.01423486,
      "auxiliary_loss_mlp": 0.01080816,
      "balance_loss_clip": 1.12266207,
      "balance_loss_mlp": 1.0541724,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.04292154266413,
      "language_loss": 0.79536057,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82040358,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.6251282691955566
    },
    {
      "auxiliary_loss_clip": 0.01419599,
      "auxiliary_loss_mlp": 0.01093269,
      "balance_loss_clip": 1.12541676,
      "balance_loss_mlp": 1.07022619,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 1.8910149269431404,
      "language_loss": 0.82591581,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85104448,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.669220209121704
    },
    {
      "auxiliary_loss_clip": 0.0141989,
      "auxiliary_loss_mlp": 0.0107794,
      "balance_loss_clip": 1.12449265,
      "balance_loss_mlp": 1.05397904,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.3092526054388736,
      "language_loss": 0.96695596,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.9919343,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.623619556427002
    },
    {
      "auxiliary_loss_clip": 0.01416412,
      "auxiliary_loss_mlp": 0.01083477,
      "balance_loss_clip": 1.11936665,
      "balance_loss_mlp": 1.05895591,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.2075962788286128,
      "language_loss": 0.84885669,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87385559,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.7622222900390625
    },
    {
      "auxiliary_loss_clip": 0.01413827,
      "auxiliary_loss_mlp": 0.0107783,
      "balance_loss_clip": 1.12257111,
      "balance_loss_mlp": 1.05656314,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.3834169292754828,
      "language_loss": 0.82816744,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85308397,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.720268964767456
    },
    {
      "auxiliary_loss_clip": 0.0140717,
      "auxiliary_loss_mlp": 0.01086382,
      "balance_loss_clip": 1.11592627,
      "balance_loss_mlp": 1.0618366,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.736017556514073,
      "language_loss": 0.86121118,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88614666,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 2.645937442779541
    },
    {
      "auxiliary_loss_clip": 0.01413345,
      "auxiliary_loss_mlp": 0.01092429,
      "balance_loss_clip": 1.12290335,
      "balance_loss_mlp": 1.06733537,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 1.9999979770646303,
      "language_loss": 0.97531509,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00037289,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 2.626147508621216
    },
    {
      "auxiliary_loss_clip": 0.01412866,
      "auxiliary_loss_mlp": 0.01069715,
      "balance_loss_clip": 1.12060022,
      "balance_loss_mlp": 1.04794765,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.9852801754529614,
      "language_loss": 0.84839594,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87322176,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.7315144538879395
    },
    {
      "auxiliary_loss_clip": 0.0141377,
      "auxiliary_loss_mlp": 0.01068186,
      "balance_loss_clip": 1.12393391,
      "balance_loss_mlp": 1.04558349,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.603236882687938,
      "language_loss": 0.95458078,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.9794004,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.6207754611968994
    },
    {
      "auxiliary_loss_clip": 0.01402327,
      "auxiliary_loss_mlp": 0.01087886,
      "balance_loss_clip": 1.11915994,
      "balance_loss_mlp": 1.06537914,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 2.076543221853463,
      "language_loss": 0.90547699,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93037915,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.6552908420562744
    },
    {
      "auxiliary_loss_clip": 0.01411356,
      "auxiliary_loss_mlp": 0.01080785,
      "balance_loss_clip": 1.12141085,
      "balance_loss_mlp": 1.05751562,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 3.36074189833122,
      "language_loss": 0.89321631,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91813767,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 3.638062000274658
    },
    {
      "auxiliary_loss_clip": 0.01401463,
      "auxiliary_loss_mlp": 0.01068077,
      "balance_loss_clip": 1.11628413,
      "balance_loss_mlp": 1.04582047,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 2.548116998661888,
      "language_loss": 0.72646332,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75115877,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 4.58662223815918
    },
    {
      "auxiliary_loss_clip": 0.0141354,
      "auxiliary_loss_mlp": 0.01087477,
      "balance_loss_clip": 1.12040341,
      "balance_loss_mlp": 1.06243157,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.23185480142561,
      "language_loss": 0.97783315,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00284338,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 3.546983242034912
    },
    {
      "auxiliary_loss_clip": 0.01407434,
      "auxiliary_loss_mlp": 0.010854,
      "balance_loss_clip": 1.11786115,
      "balance_loss_mlp": 1.06142735,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 4.734647821718854,
      "language_loss": 0.93429899,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.95922726,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.674525499343872
    },
    {
      "auxiliary_loss_clip": 0.0140861,
      "auxiliary_loss_mlp": 0.01079262,
      "balance_loss_clip": 1.11922789,
      "balance_loss_mlp": 1.05408525,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 3.618281782804491,
      "language_loss": 0.8936494,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91852814,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.647902011871338
    },
    {
      "auxiliary_loss_clip": 0.01398326,
      "auxiliary_loss_mlp": 0.01071572,
      "balance_loss_clip": 1.11778677,
      "balance_loss_mlp": 1.04994702,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.9771838902788224,
      "language_loss": 0.94191909,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96661806,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.6629064083099365
    },
    {
      "auxiliary_loss_clip": 0.01405654,
      "auxiliary_loss_mlp": 0.0107426,
      "balance_loss_clip": 1.11785674,
      "balance_loss_mlp": 1.05261111,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.2290040140750844,
      "language_loss": 0.90224719,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.9270463,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.7205333709716797
    },
    {
      "auxiliary_loss_clip": 0.01399244,
      "auxiliary_loss_mlp": 0.01084624,
      "balance_loss_clip": 1.11616015,
      "balance_loss_mlp": 1.06135464,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 2.6958643056982496,
      "language_loss": 0.82138848,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84622717,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.662031412124634
    },
    {
      "auxiliary_loss_clip": 0.01398818,
      "auxiliary_loss_mlp": 0.01087604,
      "balance_loss_clip": 1.1144197,
      "balance_loss_mlp": 1.06663537,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 3.427104500276467,
      "language_loss": 0.80667973,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83154392,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.6834962368011475
    },
    {
      "auxiliary_loss_clip": 0.01400538,
      "auxiliary_loss_mlp": 0.01084193,
      "balance_loss_clip": 1.11531711,
      "balance_loss_mlp": 1.06130457,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 2.0672319802826067,
      "language_loss": 0.85925961,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88410687,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.71178936958313
    },
    {
      "auxiliary_loss_clip": 0.01393013,
      "auxiliary_loss_mlp": 0.01076538,
      "balance_loss_clip": 1.10991037,
      "balance_loss_mlp": 1.05281484,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 4.35134768194689,
      "language_loss": 0.77160919,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.7963047,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.68691086769104
    },
    {
      "auxiliary_loss_clip": 0.01394522,
      "auxiliary_loss_mlp": 0.01059641,
      "balance_loss_clip": 1.11576831,
      "balance_loss_mlp": 1.0388031,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.184068840313121,
      "language_loss": 0.79333627,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81787789,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.8756747245788574
    },
    {
      "auxiliary_loss_clip": 0.01392187,
      "auxiliary_loss_mlp": 0.01073692,
      "balance_loss_clip": 1.1137352,
      "balance_loss_mlp": 1.05236518,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 2.007606190693602,
      "language_loss": 0.84279692,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86745566,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.673184871673584
    },
    {
      "auxiliary_loss_clip": 0.01391013,
      "auxiliary_loss_mlp": 0.01069745,
      "balance_loss_clip": 1.1123867,
      "balance_loss_mlp": 1.04964685,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.1806144909871508,
      "language_loss": 0.8834874,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.908095,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.6920435428619385
    },
    {
      "auxiliary_loss_clip": 0.01401217,
      "auxiliary_loss_mlp": 0.01070028,
      "balance_loss_clip": 1.11887467,
      "balance_loss_mlp": 1.04871368,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.358847321008234,
      "language_loss": 0.78403735,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80874979,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.689814567565918
    },
    {
      "auxiliary_loss_clip": 0.01396222,
      "auxiliary_loss_mlp": 0.01084775,
      "balance_loss_clip": 1.11553764,
      "balance_loss_mlp": 1.06263757,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 3.4317364132981667,
      "language_loss": 0.89355397,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.91836393,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.6424171924591064
    },
    {
      "auxiliary_loss_clip": 0.01379838,
      "auxiliary_loss_mlp": 0.01010552,
      "balance_loss_clip": 1.15630341,
      "balance_loss_mlp": 1.00025225,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0159502836701655,
      "language_loss": 0.63152856,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.6554324,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.1306023597717285
    },
    {
      "auxiliary_loss_clip": 0.01386514,
      "auxiliary_loss_mlp": 0.01077195,
      "balance_loss_clip": 1.11263275,
      "balance_loss_mlp": 1.05728722,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 1.906658819142814,
      "language_loss": 0.84434599,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86898309,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.636537790298462
    },
    {
      "auxiliary_loss_clip": 0.01388319,
      "auxiliary_loss_mlp": 0.01078729,
      "balance_loss_clip": 1.11199331,
      "balance_loss_mlp": 1.05613875,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 1.989343222733109,
      "language_loss": 0.87750685,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90217727,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.6776623725891113
    },
    {
      "auxiliary_loss_clip": 0.01383522,
      "auxiliary_loss_mlp": 0.01067728,
      "balance_loss_clip": 1.10864091,
      "balance_loss_mlp": 1.04742646,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.2018456875978254,
      "language_loss": 0.85914898,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88366145,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.661226272583008
    },
    {
      "auxiliary_loss_clip": 0.01386136,
      "auxiliary_loss_mlp": 0.01078622,
      "balance_loss_clip": 1.11248362,
      "balance_loss_mlp": 1.05817807,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.382623636812419,
      "language_loss": 0.82871538,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85336298,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.774233102798462
    },
    {
      "auxiliary_loss_clip": 0.01383725,
      "auxiliary_loss_mlp": 0.01082907,
      "balance_loss_clip": 1.11079168,
      "balance_loss_mlp": 1.06371427,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 1.9907092264506323,
      "language_loss": 0.84215784,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.86682409,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 2.7326555252075195
    },
    {
      "auxiliary_loss_clip": 0.01389418,
      "auxiliary_loss_mlp": 0.01082323,
      "balance_loss_clip": 1.11378932,
      "balance_loss_mlp": 1.06179476,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 3.7034139122832097,
      "language_loss": 0.78853214,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81324959,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 2.663106679916382
    },
    {
      "auxiliary_loss_clip": 0.01384742,
      "auxiliary_loss_mlp": 0.01074077,
      "balance_loss_clip": 1.11652112,
      "balance_loss_mlp": 1.05559909,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 6.4131803047711875,
      "language_loss": 0.75847393,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.7830621,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 2.6611111164093018
    },
    {
      "auxiliary_loss_clip": 0.01386568,
      "auxiliary_loss_mlp": 0.01073417,
      "balance_loss_clip": 1.10970569,
      "balance_loss_mlp": 1.05172086,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.393967335907883,
      "language_loss": 0.87030423,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89490402,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.6761844158172607
    },
    {
      "auxiliary_loss_clip": 0.01381749,
      "auxiliary_loss_mlp": 0.01077005,
      "balance_loss_clip": 1.10964417,
      "balance_loss_mlp": 1.05572581,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 2.6775285041871095,
      "language_loss": 0.84959745,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87418497,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.6038012504577637
    },
    {
      "auxiliary_loss_clip": 0.0137854,
      "auxiliary_loss_mlp": 0.01063618,
      "balance_loss_clip": 1.10763299,
      "balance_loss_mlp": 1.04345989,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.495796561357006,
      "language_loss": 0.87425739,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.89867902,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.63431978225708
    },
    {
      "auxiliary_loss_clip": 0.01378135,
      "auxiliary_loss_mlp": 0.01074459,
      "balance_loss_clip": 1.10896719,
      "balance_loss_mlp": 1.05613685,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 1.9072920912672573,
      "language_loss": 0.77093232,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79545832,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 3.5513834953308105
    },
    {
      "auxiliary_loss_clip": 0.01378583,
      "auxiliary_loss_mlp": 0.01065511,
      "balance_loss_clip": 1.10648072,
      "balance_loss_mlp": 1.04658031,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 2.374026339695792,
      "language_loss": 0.838157,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.86259794,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 3.6180179119110107
    },
    {
      "auxiliary_loss_clip": 0.01373788,
      "auxiliary_loss_mlp": 0.01059434,
      "balance_loss_clip": 1.10345328,
      "balance_loss_mlp": 1.03922832,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 2.770924270405293,
      "language_loss": 0.9146018,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93893403,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 3.5998425483703613
    },
    {
      "auxiliary_loss_clip": 0.01368218,
      "auxiliary_loss_mlp": 0.01051768,
      "balance_loss_clip": 1.10383511,
      "balance_loss_mlp": 1.03432786,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.1146705973033573,
      "language_loss": 0.9118917,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.9360916,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.7363691329956055
    },
    {
      "auxiliary_loss_clip": 0.01376773,
      "auxiliary_loss_mlp": 0.01071687,
      "balance_loss_clip": 1.10947883,
      "balance_loss_mlp": 1.05349541,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.2895798881932454,
      "language_loss": 0.9185127,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94299728,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.6166985034942627
    },
    {
      "auxiliary_loss_clip": 0.0137375,
      "auxiliary_loss_mlp": 0.01064197,
      "balance_loss_clip": 1.10604668,
      "balance_loss_mlp": 1.04545736,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 2.1563036030267315,
      "language_loss": 0.85818422,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88256371,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.6976935863494873
    },
    {
      "auxiliary_loss_clip": 0.01373886,
      "auxiliary_loss_mlp": 0.0106247,
      "balance_loss_clip": 1.10782361,
      "balance_loss_mlp": 1.04339659,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 3.4307858988896127,
      "language_loss": 0.91317272,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.9375363,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.681546449661255
    },
    {
      "auxiliary_loss_clip": 0.01369867,
      "auxiliary_loss_mlp": 0.01063824,
      "balance_loss_clip": 1.10464907,
      "balance_loss_mlp": 1.04519176,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.1679927408212807,
      "language_loss": 0.92542994,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.94976676,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.7179112434387207
    },
    {
      "auxiliary_loss_clip": 0.01367564,
      "auxiliary_loss_mlp": 0.0105961,
      "balance_loss_clip": 1.10164809,
      "balance_loss_mlp": 1.04115641,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.433513233412843,
      "language_loss": 0.91383779,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93810952,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.6805977821350098
    },
    {
      "auxiliary_loss_clip": 0.01369184,
      "auxiliary_loss_mlp": 0.01077921,
      "balance_loss_clip": 1.10497057,
      "balance_loss_mlp": 1.05928898,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.235311703488801,
      "language_loss": 0.77937853,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80384964,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.6530356407165527
    },
    {
      "auxiliary_loss_clip": 0.01367071,
      "auxiliary_loss_mlp": 0.01060509,
      "balance_loss_clip": 1.10101748,
      "balance_loss_mlp": 1.04174566,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 1.8955822897279004,
      "language_loss": 0.86534166,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.88961756,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.705204486846924
    },
    {
      "auxiliary_loss_clip": 0.01370116,
      "auxiliary_loss_mlp": 0.0105991,
      "balance_loss_clip": 1.10590577,
      "balance_loss_mlp": 1.04216003,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.251102645996209,
      "language_loss": 0.89939153,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92369181,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.634722948074341
    },
    {
      "auxiliary_loss_clip": 0.01364887,
      "auxiliary_loss_mlp": 0.01057174,
      "balance_loss_clip": 1.10357642,
      "balance_loss_mlp": 1.03847003,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.8070800029729175,
      "language_loss": 0.92108953,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94531012,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.688753366470337
    },
    {
      "auxiliary_loss_clip": 0.01365801,
      "auxiliary_loss_mlp": 0.01062348,
      "balance_loss_clip": 1.10204852,
      "balance_loss_mlp": 1.04269052,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 2.277135267139994,
      "language_loss": 0.85120785,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87548935,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.693368673324585
    },
    {
      "auxiliary_loss_clip": 0.01365491,
      "auxiliary_loss_mlp": 0.01066959,
      "balance_loss_clip": 1.10181797,
      "balance_loss_mlp": 1.0470866,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.838632330886914,
      "language_loss": 0.75272655,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77705097,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.7800662517547607
    },
    {
      "auxiliary_loss_clip": 0.01367373,
      "auxiliary_loss_mlp": 0.01068578,
      "balance_loss_clip": 1.10348105,
      "balance_loss_mlp": 1.04852688,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.420366404928414,
      "language_loss": 0.8737058,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89806533,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.651517868041992
    },
    {
      "auxiliary_loss_clip": 0.01373536,
      "auxiliary_loss_mlp": 0.01074056,
      "balance_loss_clip": 1.10643327,
      "balance_loss_mlp": 1.05197883,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 2.9736684429182496,
      "language_loss": 0.7390542,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76353014,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.867081880569458
    },
    {
      "auxiliary_loss_clip": 0.01364424,
      "auxiliary_loss_mlp": 0.01075138,
      "balance_loss_clip": 1.10194242,
      "balance_loss_mlp": 1.05509901,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 4.293513145013598,
      "language_loss": 0.81988275,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84427834,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.653350353240967
    },
    {
      "auxiliary_loss_clip": 0.01362251,
      "auxiliary_loss_mlp": 0.01065617,
      "balance_loss_clip": 1.10103858,
      "balance_loss_mlp": 1.04648376,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.850330587888969,
      "language_loss": 0.88296843,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90724719,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.645014762878418
    },
    {
      "auxiliary_loss_clip": 0.01359743,
      "auxiliary_loss_mlp": 0.01069326,
      "balance_loss_clip": 1.10073376,
      "balance_loss_mlp": 1.051314,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.664993858927401,
      "language_loss": 0.84170055,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86599123,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.7070493698120117
    },
    {
      "auxiliary_loss_clip": 0.0136651,
      "auxiliary_loss_mlp": 0.01066381,
      "balance_loss_clip": 1.10491037,
      "balance_loss_mlp": 1.04830933,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.6674539493324043,
      "language_loss": 0.76309788,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78742671,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 2.606391191482544
    },
    {
      "auxiliary_loss_clip": 0.01364329,
      "auxiliary_loss_mlp": 0.01065279,
      "balance_loss_clip": 1.10071397,
      "balance_loss_mlp": 1.0464679,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 2.116930120863402,
      "language_loss": 0.95037812,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97467422,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 2.6931569576263428
    },
    {
      "auxiliary_loss_clip": 0.01360465,
      "auxiliary_loss_mlp": 0.01071059,
      "balance_loss_clip": 1.10016656,
      "balance_loss_mlp": 1.0520097,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 3.604700529427064,
      "language_loss": 0.79780233,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82211757,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 2.671673536300659
    },
    {
      "auxiliary_loss_clip": 0.01357704,
      "auxiliary_loss_mlp": 0.01078604,
      "balance_loss_clip": 1.09905565,
      "balance_loss_mlp": 1.05907798,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 2.0697322434705496,
      "language_loss": 0.84549516,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86985826,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.755659341812134
    },
    {
      "auxiliary_loss_clip": 0.01359483,
      "auxiliary_loss_mlp": 0.01071795,
      "balance_loss_clip": 1.09800076,
      "balance_loss_mlp": 1.05353212,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 9.575986312281918,
      "language_loss": 0.77540755,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79972029,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.633054256439209
    },
    {
      "auxiliary_loss_clip": 0.01331255,
      "auxiliary_loss_mlp": 0.01014748,
      "balance_loss_clip": 1.12410319,
      "balance_loss_mlp": 1.00215995,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 1.0462902601954271,
      "language_loss": 0.59254444,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61600453,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.1935808658599854
    },
    {
      "auxiliary_loss_clip": 0.01357939,
      "auxiliary_loss_mlp": 0.01059881,
      "balance_loss_clip": 1.10045946,
      "balance_loss_mlp": 1.04074788,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.2307731563069306,
      "language_loss": 0.84313738,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86731565,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.686095714569092
    },
    {
      "auxiliary_loss_clip": 0.01359612,
      "auxiliary_loss_mlp": 0.01059018,
      "balance_loss_clip": 1.10062838,
      "balance_loss_mlp": 1.04069602,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 4.054555335780885,
      "language_loss": 0.79754984,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.8217361,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 3.6011924743652344
    },
    {
      "auxiliary_loss_clip": 0.01361893,
      "auxiliary_loss_mlp": 0.01070685,
      "balance_loss_clip": 1.10181224,
      "balance_loss_mlp": 1.05157566,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.575113502513573,
      "language_loss": 0.90636796,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93069375,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 3.5856521129608154
    },
    {
      "auxiliary_loss_clip": 0.01357882,
      "auxiliary_loss_mlp": 0.01072944,
      "balance_loss_clip": 1.09993994,
      "balance_loss_mlp": 1.05427563,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.262505751818884,
      "language_loss": 0.86152899,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.8858372,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 4.668522596359253
    },
    {
      "auxiliary_loss_clip": 0.01355254,
      "auxiliary_loss_mlp": 0.01060727,
      "balance_loss_clip": 1.09643459,
      "balance_loss_mlp": 1.04056907,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.747462940185415,
      "language_loss": 0.86555052,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.88971037,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.6205358505249023
    },
    {
      "auxiliary_loss_clip": 0.01357596,
      "auxiliary_loss_mlp": 0.01062906,
      "balance_loss_clip": 1.09816861,
      "balance_loss_mlp": 1.04324794,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.340544728925376,
      "language_loss": 0.81906688,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84327191,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.6592037677764893
    },
    {
      "auxiliary_loss_clip": 0.01352481,
      "auxiliary_loss_mlp": 0.01067112,
      "balance_loss_clip": 1.0976584,
      "balance_loss_mlp": 1.04921842,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.3369452642544193,
      "language_loss": 0.81593454,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84013045,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.6966989040374756
    },
    {
      "auxiliary_loss_clip": 0.01356887,
      "auxiliary_loss_mlp": 0.01067832,
      "balance_loss_clip": 1.09934759,
      "balance_loss_mlp": 1.04865146,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.068844001710396,
      "language_loss": 0.84428579,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86853302,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.675657272338867
    },
    {
      "auxiliary_loss_clip": 0.01363651,
      "auxiliary_loss_mlp": 0.01073772,
      "balance_loss_clip": 1.1005547,
      "balance_loss_mlp": 1.05358994,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.4851558305858243,
      "language_loss": 0.9385727,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96294689,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.7684743404388428
    },
    {
      "auxiliary_loss_clip": 0.01354032,
      "auxiliary_loss_mlp": 0.01064716,
      "balance_loss_clip": 1.09714627,
      "balance_loss_mlp": 1.04462957,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 2.548779294593831,
      "language_loss": 0.81394458,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83813202,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.5921859741210938
    },
    {
      "auxiliary_loss_clip": 0.01354497,
      "auxiliary_loss_mlp": 0.01069968,
      "balance_loss_clip": 1.09738374,
      "balance_loss_mlp": 1.0504055,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 1.9249940901642526,
      "language_loss": 0.8905946,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91483927,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.654057502746582
    },
    {
      "auxiliary_loss_clip": 0.01352431,
      "auxiliary_loss_mlp": 0.01066959,
      "balance_loss_clip": 1.09973919,
      "balance_loss_mlp": 1.0497334,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.16588566888919,
      "language_loss": 0.87902546,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90321934,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.6269009113311768
    },
    {
      "auxiliary_loss_clip": 0.01354789,
      "auxiliary_loss_mlp": 0.01063644,
      "balance_loss_clip": 1.09917998,
      "balance_loss_mlp": 1.04497552,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 2.0708871940912252,
      "language_loss": 0.87233692,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89652121,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.6205852031707764
    },
    {
      "auxiliary_loss_clip": 0.01354282,
      "auxiliary_loss_mlp": 0.01053782,
      "balance_loss_clip": 1.09663212,
      "balance_loss_mlp": 1.03462458,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.60391994315447,
      "language_loss": 0.91361213,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93769276,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.735868453979492
    },
    {
      "auxiliary_loss_clip": 0.01352619,
      "auxiliary_loss_mlp": 0.01065452,
      "balance_loss_clip": 1.09936929,
      "balance_loss_mlp": 1.04790497,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.8297861373211273,
      "language_loss": 0.92850208,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95268285,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.696018695831299
    },
    {
      "auxiliary_loss_clip": 0.01358085,
      "auxiliary_loss_mlp": 0.01062134,
      "balance_loss_clip": 1.10043812,
      "balance_loss_mlp": 1.04184437,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 4.4996990311052,
      "language_loss": 0.84672415,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87092626,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.61064076423645
    },
    {
      "auxiliary_loss_clip": 0.01354163,
      "auxiliary_loss_mlp": 0.01061786,
      "balance_loss_clip": 1.0980289,
      "balance_loss_mlp": 1.04396439,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.083585361564744,
      "language_loss": 0.89728773,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92144728,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.6923882961273193
    },
    {
      "auxiliary_loss_clip": 0.01361912,
      "auxiliary_loss_mlp": 0.01063316,
      "balance_loss_clip": 1.10375726,
      "balance_loss_mlp": 1.04573226,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 3.0242895489334862,
      "language_loss": 0.844657,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86890924,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.6791794300079346
    },
    {
      "auxiliary_loss_clip": 0.0135801,
      "auxiliary_loss_mlp": 0.01068263,
      "balance_loss_clip": 1.10092187,
      "balance_loss_mlp": 1.0493803,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.6561169552752013,
      "language_loss": 0.80429709,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82855988,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.743938446044922
    },
    {
      "auxiliary_loss_clip": 0.01348173,
      "auxiliary_loss_mlp": 0.01066518,
      "balance_loss_clip": 1.09802866,
      "balance_loss_mlp": 1.04914892,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.064880463382122,
      "language_loss": 0.82981849,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.8539654,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.7443132400512695
    },
    {
      "auxiliary_loss_clip": 0.01352465,
      "auxiliary_loss_mlp": 0.01060707,
      "balance_loss_clip": 1.09897614,
      "balance_loss_mlp": 1.04309952,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 1.8129454474897444,
      "language_loss": 0.73942089,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76355255,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.6956100463867188
    },
    {
      "auxiliary_loss_clip": 0.013521,
      "auxiliary_loss_mlp": 0.01075438,
      "balance_loss_clip": 1.09554219,
      "balance_loss_mlp": 1.05579233,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 2.4174399250329057,
      "language_loss": 0.85321099,
      "learning_rate": 4e-06,
      "loss": 0.87748635,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 2.6513779163360596
    },
    {
      "auxiliary_loss_clip": 0.01354516,
      "auxiliary_loss_mlp": 0.01057774,
      "balance_loss_clip": 1.10107183,
      "balance_loss_mlp": 1.04033399,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 3.8061125949712875,
      "language_loss": 0.88220608,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.9063291,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 2.665321111679077
    },
    {
      "auxiliary_loss_clip": 0.01345727,
      "auxiliary_loss_mlp": 0.01058129,
      "balance_loss_clip": 1.09419227,
      "balance_loss_mlp": 1.0402478,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.628628832772604,
      "language_loss": 0.89016283,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91420144,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 2.6890878677368164
    },
    {
      "auxiliary_loss_clip": 0.01345059,
      "auxiliary_loss_mlp": 0.01061791,
      "balance_loss_clip": 1.09226322,
      "balance_loss_mlp": 1.04420722,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.4119057515947033,
      "language_loss": 0.85070419,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87477267,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.696263551712036
    },
    {
      "auxiliary_loss_clip": 0.01355495,
      "auxiliary_loss_mlp": 0.01064593,
      "balance_loss_clip": 1.10141993,
      "balance_loss_mlp": 1.04652119,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 2.742706754285738,
      "language_loss": 0.96247649,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98667741,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.6853444576263428
    },
    {
      "auxiliary_loss_clip": 0.01348104,
      "auxiliary_loss_mlp": 0.01068997,
      "balance_loss_clip": 1.09486246,
      "balance_loss_mlp": 1.05138958,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.8766098525292152,
      "language_loss": 0.8760854,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.9002564,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.6185567378997803
    },
    {
      "auxiliary_loss_clip": 0.01351476,
      "auxiliary_loss_mlp": 0.01060956,
      "balance_loss_clip": 1.09481359,
      "balance_loss_mlp": 1.04157281,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.4856886593539387,
      "language_loss": 0.82193565,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84605992,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.7151546478271484
    },
    {
      "auxiliary_loss_clip": 0.01347594,
      "auxiliary_loss_mlp": 0.01063203,
      "balance_loss_clip": 1.09493303,
      "balance_loss_mlp": 1.04404616,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 3.265785152010117,
      "language_loss": 0.85554063,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.87964857,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.666616439819336
    },
    {
      "auxiliary_loss_clip": 0.01285835,
      "auxiliary_loss_mlp": 0.0101626,
      "balance_loss_clip": 1.09199858,
      "balance_loss_mlp": 1.00572157,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0188693539212086,
      "language_loss": 0.53489822,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55791914,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 5.028972864151001
    },
    {
      "auxiliary_loss_clip": 0.01347126,
      "auxiliary_loss_mlp": 0.01061314,
      "balance_loss_clip": 1.0959357,
      "balance_loss_mlp": 1.04351592,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.1279509160464833,
      "language_loss": 0.86827815,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89236253,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.668921709060669
    },
    {
      "auxiliary_loss_clip": 0.01345665,
      "auxiliary_loss_mlp": 0.01064613,
      "balance_loss_clip": 1.0968883,
      "balance_loss_mlp": 1.0466603,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 1.9382015146124538,
      "language_loss": 0.82008654,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84418929,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.6451048851013184
    },
    {
      "auxiliary_loss_clip": 0.01340598,
      "auxiliary_loss_mlp": 0.01066014,
      "balance_loss_clip": 1.09189963,
      "balance_loss_mlp": 1.04778647,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 3.6346337232871293,
      "language_loss": 0.97967792,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00374389,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 3.501112937927246
    },
    {
      "auxiliary_loss_clip": 0.01343675,
      "auxiliary_loss_mlp": 0.01069383,
      "balance_loss_clip": 1.09612417,
      "balance_loss_mlp": 1.04999971,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.2181395190987305,
      "language_loss": 0.86102366,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88515419,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.6446285247802734
    },
    {
      "auxiliary_loss_clip": 0.01341156,
      "auxiliary_loss_mlp": 0.01065902,
      "balance_loss_clip": 1.09246385,
      "balance_loss_mlp": 1.04712617,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.72269199446704,
      "language_loss": 0.80228722,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82635784,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.75152325630188
    },
    {
      "auxiliary_loss_clip": 0.01333657,
      "auxiliary_loss_mlp": 0.01059935,
      "balance_loss_clip": 1.08853149,
      "balance_loss_mlp": 1.0417552,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.207937010374817,
      "language_loss": 0.89598477,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91992068,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.654127597808838
    },
    {
      "auxiliary_loss_clip": 0.01337352,
      "auxiliary_loss_mlp": 0.01058409,
      "balance_loss_clip": 1.09224653,
      "balance_loss_mlp": 1.04053998,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 3.5865293237919333,
      "language_loss": 0.80227399,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.8262316,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.6389567852020264
    },
    {
      "auxiliary_loss_clip": 0.0133813,
      "auxiliary_loss_mlp": 0.0106424,
      "balance_loss_clip": 1.09108377,
      "balance_loss_mlp": 1.04653728,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 3.3565986404397776,
      "language_loss": 0.86876738,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89279109,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.6456072330474854
    },
    {
      "auxiliary_loss_clip": 0.01342311,
      "auxiliary_loss_mlp": 0.01060633,
      "balance_loss_clip": 1.09454763,
      "balance_loss_mlp": 1.04277539,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 4.02627595643558,
      "language_loss": 0.78859806,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81262743,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.593566417694092
    },
    {
      "auxiliary_loss_clip": 0.01339182,
      "auxiliary_loss_mlp": 0.01064848,
      "balance_loss_clip": 1.0932548,
      "balance_loss_mlp": 1.04700267,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 1.9896087287395388,
      "language_loss": 0.79677999,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.82082033,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.7544918060302734
    },
    {
      "auxiliary_loss_clip": 0.01345842,
      "auxiliary_loss_mlp": 0.01068865,
      "balance_loss_clip": 1.09558642,
      "balance_loss_mlp": 1.05112672,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.643542538723376,
      "language_loss": 0.84207702,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86622405,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.56972336769104
    },
    {
      "auxiliary_loss_clip": 0.01344757,
      "auxiliary_loss_mlp": 0.0106403,
      "balance_loss_clip": 1.09663057,
      "balance_loss_mlp": 1.04482555,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.4534880626387707,
      "language_loss": 0.94807452,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97216243,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.6513397693634033
    },
    {
      "auxiliary_loss_clip": 0.01341772,
      "auxiliary_loss_mlp": 0.01063797,
      "balance_loss_clip": 1.09508157,
      "balance_loss_mlp": 1.04579616,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.0377027809290302,
      "language_loss": 0.89693892,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92099464,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.636606216430664
    },
    {
      "auxiliary_loss_clip": 0.01339053,
      "auxiliary_loss_mlp": 0.01065958,
      "balance_loss_clip": 1.09369779,
      "balance_loss_mlp": 1.04795766,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 2.3228806817972876,
      "language_loss": 0.86745679,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89150691,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.694326400756836
    },
    {
      "auxiliary_loss_clip": 0.01336353,
      "auxiliary_loss_mlp": 0.01067531,
      "balance_loss_clip": 1.08901715,
      "balance_loss_mlp": 1.04935122,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 1.9736931346646238,
      "language_loss": 0.62966782,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65370667,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.8298721313476562
    },
    {
      "auxiliary_loss_clip": 0.01332359,
      "auxiliary_loss_mlp": 0.01053844,
      "balance_loss_clip": 1.08806765,
      "balance_loss_mlp": 1.03498554,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.4663792882843243,
      "language_loss": 0.75802612,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78188813,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.664553165435791
    },
    {
      "auxiliary_loss_clip": 0.01339466,
      "auxiliary_loss_mlp": 0.01050562,
      "balance_loss_clip": 1.09368718,
      "balance_loss_mlp": 1.03262138,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.3679241191490163,
      "language_loss": 0.89904529,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.92294562,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.608426094055176
    },
    {
      "auxiliary_loss_clip": 0.01337114,
      "auxiliary_loss_mlp": 0.01066359,
      "balance_loss_clip": 1.09266794,
      "balance_loss_mlp": 1.04882383,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.4048834013068165,
      "language_loss": 0.92506915,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94910383,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 2.7356748580932617
    },
    {
      "auxiliary_loss_clip": 0.01331655,
      "auxiliary_loss_mlp": 0.01059627,
      "balance_loss_clip": 1.08921707,
      "balance_loss_mlp": 1.04130435,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.2744931547478227,
      "language_loss": 1.00265861,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02657151,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 2.646152973175049
    },
    {
      "auxiliary_loss_clip": 0.01332052,
      "auxiliary_loss_mlp": 0.01041746,
      "balance_loss_clip": 1.08803415,
      "balance_loss_mlp": 1.02506852,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 2.0418793200243366,
      "language_loss": 0.79112136,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81485933,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 2.6789443492889404
    },
    {
      "auxiliary_loss_clip": 0.01333432,
      "auxiliary_loss_mlp": 0.01056295,
      "balance_loss_clip": 1.08812726,
      "balance_loss_mlp": 1.03761494,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.5702130587044745,
      "language_loss": 0.8704074,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89430463,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.642270803451538
    },
    {
      "auxiliary_loss_clip": 0.01336054,
      "auxiliary_loss_mlp": 0.01063567,
      "balance_loss_clip": 1.09243488,
      "balance_loss_mlp": 1.04523253,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 2.3005111456120444,
      "language_loss": 0.79458082,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.81857705,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.6242928504943848
    },
    {
      "auxiliary_loss_clip": 0.01334441,
      "auxiliary_loss_mlp": 0.01056598,
      "balance_loss_clip": 1.09180892,
      "balance_loss_mlp": 1.03912234,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 3.543021106992283,
      "language_loss": 0.784217,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.8081274,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.749788761138916
    },
    {
      "auxiliary_loss_clip": 0.0133362,
      "auxiliary_loss_mlp": 0.01061903,
      "balance_loss_clip": 1.09120083,
      "balance_loss_mlp": 1.04391396,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.0661259024092367,
      "language_loss": 0.82020628,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84416157,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.676923990249634
    },
    {
      "auxiliary_loss_clip": 0.01326738,
      "auxiliary_loss_mlp": 0.01059043,
      "balance_loss_clip": 1.0871588,
      "balance_loss_mlp": 1.04140019,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 2.9922016103828684,
      "language_loss": 0.92178202,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94563985,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.6474297046661377
    },
    {
      "auxiliary_loss_clip": 0.01327507,
      "auxiliary_loss_mlp": 0.01055318,
      "balance_loss_clip": 1.08766329,
      "balance_loss_mlp": 1.03789008,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.2792868647037627,
      "language_loss": 0.7424686,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.7662968,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.6428377628326416
    },
    {
      "auxiliary_loss_clip": 0.01332622,
      "auxiliary_loss_mlp": 0.01052095,
      "balance_loss_clip": 1.08711219,
      "balance_loss_mlp": 1.03341508,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.6554815892548618,
      "language_loss": 0.90955687,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93340409,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 3.500286340713501
    },
    {
      "auxiliary_loss_clip": 0.01327107,
      "auxiliary_loss_mlp": 0.01059318,
      "balance_loss_clip": 1.08753037,
      "balance_loss_mlp": 1.04250968,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 2.178941289792717,
      "language_loss": 0.7354027,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75926697,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 3.730577230453491
    },
    {
      "auxiliary_loss_clip": 0.01327652,
      "auxiliary_loss_mlp": 0.0106188,
      "balance_loss_clip": 1.08913445,
      "balance_loss_mlp": 1.04472613,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 3.4222984661636344,
      "language_loss": 0.75863558,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.7825309,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.6404659748077393
    },
    {
      "auxiliary_loss_clip": 0.01334193,
      "auxiliary_loss_mlp": 0.01061305,
      "balance_loss_clip": 1.09402323,
      "balance_loss_mlp": 1.04409099,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.581165669290499,
      "language_loss": 0.96409178,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.98804665,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 3.571047067642212
    },
    {
      "auxiliary_loss_clip": 0.01334768,
      "auxiliary_loss_mlp": 0.01066272,
      "balance_loss_clip": 1.09062171,
      "balance_loss_mlp": 1.04780686,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.8279292700614633,
      "language_loss": 0.82720733,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.85121775,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.6687564849853516
    },
    {
      "auxiliary_loss_clip": 0.01328923,
      "auxiliary_loss_mlp": 0.01056868,
      "balance_loss_clip": 1.08850718,
      "balance_loss_mlp": 1.0392251,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.8008455233558205,
      "language_loss": 0.8088991,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.832757,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.7274773120880127
    },
    {
      "auxiliary_loss_clip": 0.01332437,
      "auxiliary_loss_mlp": 0.0105388,
      "balance_loss_clip": 1.09081542,
      "balance_loss_mlp": 1.03591537,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 2.000569892453352,
      "language_loss": 0.90941226,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93327546,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.6791470050811768
    },
    {
      "auxiliary_loss_clip": 0.01276765,
      "auxiliary_loss_mlp": 0.01008357,
      "balance_loss_clip": 1.09591341,
      "balance_loss_mlp": 0.99986917,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9583401087904283,
      "language_loss": 0.61150014,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63435137,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.1288275718688965
    },
    {
      "auxiliary_loss_clip": 0.01325944,
      "auxiliary_loss_mlp": 0.01060224,
      "balance_loss_clip": 1.08583045,
      "balance_loss_mlp": 1.04221153,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 5.551285840095868,
      "language_loss": 0.92113864,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94500029,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.655733346939087
    },
    {
      "auxiliary_loss_clip": 0.0132336,
      "auxiliary_loss_mlp": 0.0105593,
      "balance_loss_clip": 1.08577931,
      "balance_loss_mlp": 1.03934753,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.550709804193853,
      "language_loss": 0.74965358,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77344644,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.728248357772827
    },
    {
      "auxiliary_loss_clip": 0.01324753,
      "auxiliary_loss_mlp": 0.01058603,
      "balance_loss_clip": 1.08703947,
      "balance_loss_mlp": 1.04211664,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.1388045945586516,
      "language_loss": 0.85892034,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88275397,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.6314165592193604
    },
    {
      "auxiliary_loss_clip": 0.01267886,
      "auxiliary_loss_mlp": 0.01010313,
      "balance_loss_clip": 1.09081364,
      "balance_loss_mlp": 1.0019207,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 0.9980189924886462,
      "language_loss": 0.69414699,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71692896,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.2039122581481934
    },
    {
      "auxiliary_loss_clip": 0.01327786,
      "auxiliary_loss_mlp": 0.01059111,
      "balance_loss_clip": 1.08852339,
      "balance_loss_mlp": 1.04190898,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 3.4903318049074348,
      "language_loss": 0.85607088,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87993985,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.600213050842285
    },
    {
      "auxiliary_loss_clip": 0.01330255,
      "auxiliary_loss_mlp": 0.01059732,
      "balance_loss_clip": 1.08855665,
      "balance_loss_mlp": 1.04273272,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.1662669361849063,
      "language_loss": 0.65487218,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.67877203,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.794111490249634
    },
    {
      "auxiliary_loss_clip": 0.01322444,
      "auxiliary_loss_mlp": 0.01060556,
      "balance_loss_clip": 1.08682275,
      "balance_loss_mlp": 1.04304433,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 2.7949730413243032,
      "language_loss": 0.90331459,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92714465,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.626791000366211
    },
    {
      "auxiliary_loss_clip": 0.01254932,
      "auxiliary_loss_mlp": 0.01009118,
      "balance_loss_clip": 1.0822041,
      "balance_loss_mlp": 1.00053453,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9172166554384111,
      "language_loss": 0.52248442,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54512501,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 2.9869449138641357
    },
    {
      "auxiliary_loss_clip": 0.01321044,
      "auxiliary_loss_mlp": 0.01055529,
      "balance_loss_clip": 1.08665299,
      "balance_loss_mlp": 1.03830338,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 2.202795272631024,
      "language_loss": 0.86571008,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88947582,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.645508289337158
    },
    {
      "auxiliary_loss_clip": 0.01328368,
      "auxiliary_loss_mlp": 0.01062542,
      "balance_loss_clip": 1.08914852,
      "balance_loss_mlp": 1.04504216,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.706364570108286,
      "language_loss": 0.89442372,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91833287,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.6218137741088867
    },
    {
      "auxiliary_loss_clip": 0.01327595,
      "auxiliary_loss_mlp": 0.01053848,
      "balance_loss_clip": 1.08820236,
      "balance_loss_mlp": 1.03572798,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.026111611535276,
      "language_loss": 0.84856296,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87237734,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 2.681464910507202
    },
    {
      "auxiliary_loss_clip": 0.0133053,
      "auxiliary_loss_mlp": 0.01060743,
      "balance_loss_clip": 1.08961689,
      "balance_loss_mlp": 1.04182434,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.131802915694188,
      "language_loss": 0.8115952,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83550793,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 2.694868564605713
    },
    {
      "auxiliary_loss_clip": 0.01329708,
      "auxiliary_loss_mlp": 0.01054128,
      "balance_loss_clip": 1.09095168,
      "balance_loss_mlp": 1.03759336,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 1.9614769604521165,
      "language_loss": 0.81224149,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.8360799,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.629507303237915
    },
    {
      "auxiliary_loss_clip": 0.01325326,
      "auxiliary_loss_mlp": 0.01064603,
      "balance_loss_clip": 1.08699393,
      "balance_loss_mlp": 1.04639935,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 2.046319979282649,
      "language_loss": 0.81710643,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.84100568,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 2.731678009033203
    },
    {
      "auxiliary_loss_clip": 0.01323078,
      "auxiliary_loss_mlp": 0.01062042,
      "balance_loss_clip": 1.08840227,
      "balance_loss_mlp": 1.04484046,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.174461019019535,
      "language_loss": 0.92519236,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94904363,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.6193978786468506
    },
    {
      "auxiliary_loss_clip": 0.01322853,
      "auxiliary_loss_mlp": 0.01053066,
      "balance_loss_clip": 1.08762741,
      "balance_loss_mlp": 1.03517246,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 3.9687354577523775,
      "language_loss": 0.97751564,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.00127482,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.6652486324310303
    },
    {
      "auxiliary_loss_clip": 0.01320033,
      "auxiliary_loss_mlp": 0.0106041,
      "balance_loss_clip": 1.08368468,
      "balance_loss_mlp": 1.0433985,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.1931193634335164,
      "language_loss": 0.86110276,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88490719,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.6377673149108887
    },
    {
      "auxiliary_loss_clip": 0.01326335,
      "auxiliary_loss_mlp": 0.01064678,
      "balance_loss_clip": 1.08863473,
      "balance_loss_mlp": 1.04666519,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 2.1394861630575805,
      "language_loss": 0.76145256,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78536272,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.7320234775543213
    },
    {
      "auxiliary_loss_clip": 0.01322528,
      "auxiliary_loss_mlp": 0.01050291,
      "balance_loss_clip": 1.08667874,
      "balance_loss_mlp": 1.03372097,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 1.8256299488942604,
      "language_loss": 0.81137216,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83510029,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.7382211685180664
    },
    {
      "auxiliary_loss_clip": 0.01320267,
      "auxiliary_loss_mlp": 0.01061337,
      "balance_loss_clip": 1.08564663,
      "balance_loss_mlp": 1.04436171,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.441766753831645,
      "language_loss": 0.82908177,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85289788,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 3.5764577388763428
    },
    {
      "auxiliary_loss_clip": 0.01321413,
      "auxiliary_loss_mlp": 0.01054376,
      "balance_loss_clip": 1.08493137,
      "balance_loss_mlp": 1.03704286,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 2.8121122611263965,
      "language_loss": 0.85889643,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88265431,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.631338119506836
    },
    {
      "auxiliary_loss_clip": 0.01329868,
      "auxiliary_loss_mlp": 0.01061099,
      "balance_loss_clip": 1.09065366,
      "balance_loss_mlp": 1.04375398,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 6.542539263757876,
      "language_loss": 0.82762027,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.8515299,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 5.360571622848511
    },
    {
      "auxiliary_loss_clip": 0.01322485,
      "auxiliary_loss_mlp": 0.0105456,
      "balance_loss_clip": 1.08775258,
      "balance_loss_mlp": 1.035725,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 2.0083122346925903,
      "language_loss": 0.82884842,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85261881,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.6103713512420654
    },
    {
      "auxiliary_loss_clip": 0.01322633,
      "auxiliary_loss_mlp": 0.01063923,
      "balance_loss_clip": 1.08699393,
      "balance_loss_mlp": 1.04619646,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 1.9862062398005393,
      "language_loss": 0.78838611,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81225169,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.660637378692627
    },
    {
      "auxiliary_loss_clip": 0.01323358,
      "auxiliary_loss_mlp": 0.01049928,
      "balance_loss_clip": 1.08812237,
      "balance_loss_mlp": 1.03178406,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 2.846393244989476,
      "language_loss": 0.83599102,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.85972381,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.6504526138305664
    },
    {
      "auxiliary_loss_clip": 0.01314339,
      "auxiliary_loss_mlp": 0.01050209,
      "balance_loss_clip": 1.08245444,
      "balance_loss_mlp": 1.03282857,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.5670041069780636,
      "language_loss": 0.8285321,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85217762,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.707681179046631
    },
    {
      "auxiliary_loss_clip": 0.01317346,
      "auxiliary_loss_mlp": 0.01058632,
      "balance_loss_clip": 1.08357811,
      "balance_loss_mlp": 1.04138255,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.059321035180556,
      "language_loss": 0.77513766,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.7988975,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.611294746398926
    },
    {
      "auxiliary_loss_clip": 0.01225896,
      "auxiliary_loss_mlp": 0.01013497,
      "balance_loss_clip": 1.06397963,
      "balance_loss_mlp": 1.00496185,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9576152171022124,
      "language_loss": 0.62355185,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64594579,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.1089565753936768
    },
    {
      "auxiliary_loss_clip": 0.0122338,
      "auxiliary_loss_mlp": 0.01010249,
      "balance_loss_clip": 1.06219864,
      "balance_loss_mlp": 1.0015707,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2148110395570102,
      "language_loss": 0.67119479,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69353116,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.070452928543091
    },
    {
      "auxiliary_loss_clip": 0.01317306,
      "auxiliary_loss_mlp": 0.01055999,
      "balance_loss_clip": 1.08664954,
      "balance_loss_mlp": 1.03946471,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 1.9002884420485444,
      "language_loss": 0.82132608,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.8450591,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.6321792602539062
    },
    {
      "auxiliary_loss_clip": 0.01316704,
      "auxiliary_loss_mlp": 0.01052433,
      "balance_loss_clip": 1.0833509,
      "balance_loss_mlp": 1.03602982,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 1.9573435549402967,
      "language_loss": 0.91016644,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.9338578,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.7064785957336426
    },
    {
      "auxiliary_loss_clip": 0.01217523,
      "auxiliary_loss_mlp": 0.01009811,
      "balance_loss_clip": 1.05853868,
      "balance_loss_mlp": 1.00160909,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9104064564221976,
      "language_loss": 0.59788215,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62015545,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.109179973602295
    },
    {
      "auxiliary_loss_clip": 0.01319369,
      "auxiliary_loss_mlp": 0.01079036,
      "balance_loss_clip": 1.08502626,
      "balance_loss_mlp": 1.06061769,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 2.043114214181247,
      "language_loss": 0.82950628,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85349035,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.634340524673462
    },
    {
      "auxiliary_loss_clip": 0.01316586,
      "auxiliary_loss_mlp": 0.01059664,
      "balance_loss_clip": 1.08335531,
      "balance_loss_mlp": 1.042045,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.269534216674422,
      "language_loss": 0.8927148,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91647732,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.579180955886841
    },
    {
      "auxiliary_loss_clip": 0.01317245,
      "auxiliary_loss_mlp": 0.01056613,
      "balance_loss_clip": 1.08402002,
      "balance_loss_mlp": 1.03836203,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.363363644953309,
      "language_loss": 0.7475729,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77131146,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.66227126121521
    },
    {
      "auxiliary_loss_clip": 0.01316906,
      "auxiliary_loss_mlp": 0.01062318,
      "balance_loss_clip": 1.08524418,
      "balance_loss_mlp": 1.04484153,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.427019886296282,
      "language_loss": 0.88260096,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90639317,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.6111576557159424
    },
    {
      "auxiliary_loss_clip": 0.01319384,
      "auxiliary_loss_mlp": 0.01063508,
      "balance_loss_clip": 1.08574867,
      "balance_loss_mlp": 1.04368377,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 1.8086801187969395,
      "language_loss": 0.90655875,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93038762,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 2.7108566761016846
    },
    {
      "auxiliary_loss_clip": 0.01315747,
      "auxiliary_loss_mlp": 0.01056383,
      "balance_loss_clip": 1.08314812,
      "balance_loss_mlp": 1.03841841,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.5433918649442715,
      "language_loss": 0.75828487,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.7820062,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 2.591304063796997
    },
    {
      "auxiliary_loss_clip": 0.01313075,
      "auxiliary_loss_mlp": 0.01051564,
      "balance_loss_clip": 1.0831126,
      "balance_loss_mlp": 1.03382564,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.0747309434354797,
      "language_loss": 0.86899698,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.89264333,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 2.6473793983459473
    },
    {
      "auxiliary_loss_clip": 0.01315445,
      "auxiliary_loss_mlp": 0.01057062,
      "balance_loss_clip": 1.08729672,
      "balance_loss_mlp": 1.04137349,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.3730617389797617,
      "language_loss": 0.82318306,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84690809,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.6395084857940674
    },
    {
      "auxiliary_loss_clip": 0.01320799,
      "auxiliary_loss_mlp": 0.01053849,
      "balance_loss_clip": 1.08513045,
      "balance_loss_mlp": 1.03555012,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.4760396282788983,
      "language_loss": 0.75801414,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78176057,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.694798469543457
    },
    {
      "auxiliary_loss_clip": 0.01315656,
      "auxiliary_loss_mlp": 0.01054666,
      "balance_loss_clip": 1.08584511,
      "balance_loss_mlp": 1.03825068,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.399283988612303,
      "language_loss": 0.84292042,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86662364,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.6890065670013428
    },
    {
      "auxiliary_loss_clip": 0.01314335,
      "auxiliary_loss_mlp": 0.01057514,
      "balance_loss_clip": 1.08649373,
      "balance_loss_mlp": 1.04086065,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.6539153605932326,
      "language_loss": 0.88988256,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.9136011,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.6084494590759277
    },
    {
      "auxiliary_loss_clip": 0.01309943,
      "auxiliary_loss_mlp": 0.01056003,
      "balance_loss_clip": 1.07929575,
      "balance_loss_mlp": 1.03878927,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.7870904342479115,
      "language_loss": 0.8683641,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89202356,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.7685904502868652
    },
    {
      "auxiliary_loss_clip": 0.0131282,
      "auxiliary_loss_mlp": 0.01051411,
      "balance_loss_clip": 1.08243704,
      "balance_loss_mlp": 1.03363705,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 4.197704678754756,
      "language_loss": 0.84957439,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87321663,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.6423499584198
    },
    {
      "auxiliary_loss_clip": 0.0131032,
      "auxiliary_loss_mlp": 0.01047427,
      "balance_loss_clip": 1.08147931,
      "balance_loss_mlp": 1.030797,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 2.1998285406677445,
      "language_loss": 0.83213568,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85571313,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.649712324142456
    },
    {
      "auxiliary_loss_clip": 0.0131341,
      "auxiliary_loss_mlp": 0.01061408,
      "balance_loss_clip": 1.08193815,
      "balance_loss_mlp": 1.04413438,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 3.9811350932902365,
      "language_loss": 0.7326144,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75636256,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 3.5663859844207764
    },
    {
      "auxiliary_loss_clip": 0.01313452,
      "auxiliary_loss_mlp": 0.01055324,
      "balance_loss_clip": 1.08230305,
      "balance_loss_mlp": 1.03806257,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.143500135514113,
      "language_loss": 0.83672321,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86041099,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.6585605144500732
    },
    {
      "auxiliary_loss_clip": 0.01308303,
      "auxiliary_loss_mlp": 0.01053374,
      "balance_loss_clip": 1.08127832,
      "balance_loss_mlp": 1.03661346,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 1.8907358160050283,
      "language_loss": 0.87915421,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.902771,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 4.5073463916778564
    },
    {
      "auxiliary_loss_clip": 0.01318965,
      "auxiliary_loss_mlp": 0.01049767,
      "balance_loss_clip": 1.09113789,
      "balance_loss_mlp": 1.03144503,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 3.0623000187086897,
      "language_loss": 0.73442966,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75811702,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.591179847717285
    },
    {
      "auxiliary_loss_clip": 0.01315941,
      "auxiliary_loss_mlp": 0.01058131,
      "balance_loss_clip": 1.08764386,
      "balance_loss_mlp": 1.04053533,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 2.607449345913317,
      "language_loss": 0.90437114,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92811179,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.6529858112335205
    },
    {
      "auxiliary_loss_clip": 0.01315458,
      "auxiliary_loss_mlp": 0.01054899,
      "balance_loss_clip": 1.08502543,
      "balance_loss_mlp": 1.03859115,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.843560316522009,
      "language_loss": 0.80004841,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82375199,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.7389798164367676
    },
    {
      "auxiliary_loss_clip": 0.01313327,
      "auxiliary_loss_mlp": 0.01055849,
      "balance_loss_clip": 1.08383965,
      "balance_loss_mlp": 1.03861129,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 1.8051094823757554,
      "language_loss": 0.7779305,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80162227,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.646439552307129
    },
    {
      "auxiliary_loss_clip": 0.01311836,
      "auxiliary_loss_mlp": 0.01060447,
      "balance_loss_clip": 1.08607864,
      "balance_loss_mlp": 1.04484296,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.8031474172232342,
      "language_loss": 0.83491027,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85863316,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.6177902221679688
    },
    {
      "auxiliary_loss_clip": 0.01312258,
      "auxiliary_loss_mlp": 0.010565,
      "balance_loss_clip": 1.08245587,
      "balance_loss_mlp": 1.03780794,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 2.8952320608397133,
      "language_loss": 0.91819549,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.94188303,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.6267104148864746
    },
    {
      "auxiliary_loss_clip": 0.0131582,
      "auxiliary_loss_mlp": 0.01055218,
      "balance_loss_clip": 1.08260393,
      "balance_loss_mlp": 1.03727698,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.333579772709652,
      "language_loss": 0.85189128,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87560171,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.6121325492858887
    },
    {
      "auxiliary_loss_clip": 0.01312425,
      "auxiliary_loss_mlp": 0.01066695,
      "balance_loss_clip": 1.08450603,
      "balance_loss_mlp": 1.05006576,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 2.044329800708247,
      "language_loss": 0.87960792,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90339911,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.5780386924743652
    },
    {
      "auxiliary_loss_clip": 0.01318832,
      "auxiliary_loss_mlp": 0.01077242,
      "balance_loss_clip": 1.08653975,
      "balance_loss_mlp": 1.05801368,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 3.133744572474356,
      "language_loss": 0.94922733,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97318804,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.6836087703704834
    },
    {
      "auxiliary_loss_clip": 0.0132078,
      "auxiliary_loss_mlp": 0.01065473,
      "balance_loss_clip": 1.08840966,
      "balance_loss_mlp": 1.04710269,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.485948764752713,
      "language_loss": 0.78873336,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.8125959,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.577904224395752
    },
    {
      "auxiliary_loss_clip": 0.01312955,
      "auxiliary_loss_mlp": 0.0105499,
      "balance_loss_clip": 1.08179879,
      "balance_loss_mlp": 1.03842008,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.970327574942098,
      "language_loss": 0.84359741,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86727691,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.647308588027954
    },
    {
      "auxiliary_loss_clip": 0.01311576,
      "auxiliary_loss_mlp": 0.01052079,
      "balance_loss_clip": 1.08524382,
      "balance_loss_mlp": 1.03623569,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.605252488616307,
      "language_loss": 0.89185435,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91549093,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.6674816608428955
    },
    {
      "auxiliary_loss_clip": 0.01312641,
      "auxiliary_loss_mlp": 0.01062828,
      "balance_loss_clip": 1.08425415,
      "balance_loss_mlp": 1.04635358,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.4466215705284893,
      "language_loss": 0.90325278,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92700744,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.60200572013855
    },
    {
      "auxiliary_loss_clip": 0.01310815,
      "auxiliary_loss_mlp": 0.01051612,
      "balance_loss_clip": 1.08172441,
      "balance_loss_mlp": 1.03576958,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.657014517057481,
      "language_loss": 0.8111124,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83473665,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.595211982727051
    },
    {
      "auxiliary_loss_clip": 0.01314083,
      "auxiliary_loss_mlp": 0.01053274,
      "balance_loss_clip": 1.08164239,
      "balance_loss_mlp": 1.03689456,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.9185416915989002,
      "language_loss": 0.81122452,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83489811,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.742190361022949
    },
    {
      "auxiliary_loss_clip": 0.0131683,
      "auxiliary_loss_mlp": 0.01064403,
      "balance_loss_clip": 1.08873057,
      "balance_loss_mlp": 1.04696238,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 2.0535848103112597,
      "language_loss": 0.83189225,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85570455,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 2.6447677612304688
    },
    {
      "auxiliary_loss_clip": 0.01310217,
      "auxiliary_loss_mlp": 0.01048492,
      "balance_loss_clip": 1.08139086,
      "balance_loss_mlp": 1.0321604,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 2.27626084064452,
      "language_loss": 0.81947988,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84306705,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 2.7742207050323486
    },
    {
      "auxiliary_loss_clip": 0.01310931,
      "auxiliary_loss_mlp": 0.01054021,
      "balance_loss_clip": 1.08550632,
      "balance_loss_mlp": 1.03774929,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.2219725283035503,
      "language_loss": 0.84653509,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.8701846,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.6039633750915527
    },
    {
      "auxiliary_loss_clip": 0.01304393,
      "auxiliary_loss_mlp": 0.01057662,
      "balance_loss_clip": 1.08121777,
      "balance_loss_mlp": 1.04180717,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.1799660197301103,
      "language_loss": 0.85997677,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88359737,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.6095361709594727
    },
    {
      "auxiliary_loss_clip": 0.0130922,
      "auxiliary_loss_mlp": 0.0106172,
      "balance_loss_clip": 1.08106232,
      "balance_loss_mlp": 1.04591298,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 3.964887499800549,
      "language_loss": 0.96038628,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98409563,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.580925464630127
    },
    {
      "auxiliary_loss_clip": 0.01308287,
      "auxiliary_loss_mlp": 0.01061612,
      "balance_loss_clip": 1.08133793,
      "balance_loss_mlp": 1.04566157,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 5.256967785860186,
      "language_loss": 0.76551932,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.78921831,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.6427736282348633
    },
    {
      "auxiliary_loss_clip": 0.01304059,
      "auxiliary_loss_mlp": 0.01067328,
      "balance_loss_clip": 1.0804348,
      "balance_loss_mlp": 1.05086541,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.7492701844704035,
      "language_loss": 0.84972513,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87343901,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.680210590362549
    },
    {
      "auxiliary_loss_clip": 0.01307478,
      "auxiliary_loss_mlp": 0.01057599,
      "balance_loss_clip": 1.08233094,
      "balance_loss_mlp": 1.04175591,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 1.850394790363539,
      "language_loss": 0.80870575,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83235657,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.5701417922973633
    },
    {
      "auxiliary_loss_clip": 0.01307622,
      "auxiliary_loss_mlp": 0.01047181,
      "balance_loss_clip": 1.08044004,
      "balance_loss_mlp": 1.03071856,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.207065665695492,
      "language_loss": 0.82233346,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84588152,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.737130641937256
    },
    {
      "auxiliary_loss_clip": 0.01308342,
      "auxiliary_loss_mlp": 0.01049558,
      "balance_loss_clip": 1.0816021,
      "balance_loss_mlp": 1.03297627,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.6186756482847207,
      "language_loss": 0.83304977,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85662878,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.6694324016571045
    },
    {
      "auxiliary_loss_clip": 0.01310889,
      "auxiliary_loss_mlp": 0.01055423,
      "balance_loss_clip": 1.08397198,
      "balance_loss_mlp": 1.03853083,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.1524240947736737,
      "language_loss": 0.8120513,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.8357144,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 3.588226795196533
    },
    {
      "auxiliary_loss_clip": 0.01305531,
      "auxiliary_loss_mlp": 0.01051474,
      "balance_loss_clip": 1.08340001,
      "balance_loss_mlp": 1.03640568,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.7165668043339968,
      "language_loss": 0.88616741,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90973753,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 3.5792667865753174
    },
    {
      "auxiliary_loss_clip": 0.01305715,
      "auxiliary_loss_mlp": 0.0104848,
      "balance_loss_clip": 1.08023787,
      "balance_loss_mlp": 1.03186226,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 1.9627943226306357,
      "language_loss": 0.79195988,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81550181,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.5831944942474365
    },
    {
      "auxiliary_loss_clip": 0.01305431,
      "auxiliary_loss_mlp": 0.01060387,
      "balance_loss_clip": 1.07966256,
      "balance_loss_mlp": 1.04419875,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 7.051017832801944,
      "language_loss": 0.89251977,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91617793,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 3.5641229152679443
    },
    {
      "auxiliary_loss_clip": 0.01306296,
      "auxiliary_loss_mlp": 0.01065157,
      "balance_loss_clip": 1.08127975,
      "balance_loss_mlp": 1.04864621,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 1.8970416606382634,
      "language_loss": 0.80172926,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82544386,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.658844232559204
    },
    {
      "auxiliary_loss_clip": 0.01300677,
      "auxiliary_loss_mlp": 0.01052587,
      "balance_loss_clip": 1.07879329,
      "balance_loss_mlp": 1.03608871,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.508309376752762,
      "language_loss": 0.7356106,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.75914323,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.6411545276641846
    },
    {
      "auxiliary_loss_clip": 0.01306657,
      "auxiliary_loss_mlp": 0.01062179,
      "balance_loss_clip": 1.08294678,
      "balance_loss_mlp": 1.04618108,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.739747329027154,
      "language_loss": 0.9283042,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.95199257,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.5536983013153076
    },
    {
      "auxiliary_loss_clip": 0.01204494,
      "auxiliary_loss_mlp": 0.01021253,
      "balance_loss_clip": 1.05611992,
      "balance_loss_mlp": 1.01457727,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 1.0017829744855573,
      "language_loss": 0.60247755,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62473506,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.081347703933716
    },
    {
      "auxiliary_loss_clip": 0.01203463,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.05454159,
      "balance_loss_mlp": 1.01737928,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.8996600841633126,
      "language_loss": 0.57743013,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.59971201,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.253336191177368
    },
    {
      "auxiliary_loss_clip": 0.01306632,
      "auxiliary_loss_mlp": 0.01059186,
      "balance_loss_clip": 1.08120143,
      "balance_loss_mlp": 1.04274654,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 1.8202450196766908,
      "language_loss": 0.77287197,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79653013,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.6211631298065186
    },
    {
      "auxiliary_loss_clip": 0.01299586,
      "auxiliary_loss_mlp": 0.01048285,
      "balance_loss_clip": 1.07825875,
      "balance_loss_mlp": 1.03237057,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.4325030103423475,
      "language_loss": 0.65552819,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.67900687,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.7020246982574463
    },
    {
      "auxiliary_loss_clip": 0.01301373,
      "auxiliary_loss_mlp": 0.01055338,
      "balance_loss_clip": 1.07912982,
      "balance_loss_mlp": 1.03919685,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.6809622214177067,
      "language_loss": 0.91118532,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.9347524,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.6193087100982666
    },
    {
      "auxiliary_loss_clip": 0.01302046,
      "auxiliary_loss_mlp": 0.01054337,
      "balance_loss_clip": 1.07907414,
      "balance_loss_mlp": 1.03850627,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 2.2038302745294382,
      "language_loss": 0.86772519,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.891289,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.6244466304779053
    },
    {
      "auxiliary_loss_clip": 0.01300774,
      "auxiliary_loss_mlp": 0.01052935,
      "balance_loss_clip": 1.08113956,
      "balance_loss_mlp": 1.03742576,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 1.9685563845491367,
      "language_loss": 0.77305698,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79659408,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.5804011821746826
    },
    {
      "auxiliary_loss_clip": 0.01310363,
      "auxiliary_loss_mlp": 0.0105653,
      "balance_loss_clip": 1.08265305,
      "balance_loss_mlp": 1.03860092,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.6993392936143676,
      "language_loss": 0.74330688,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76697576,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.6916990280151367
    },
    {
      "auxiliary_loss_clip": 0.01193883,
      "auxiliary_loss_mlp": 0.01008053,
      "balance_loss_clip": 1.04676247,
      "balance_loss_mlp": 0.99989909,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8184147306291194,
      "language_loss": 0.60050243,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62252176,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 3.2506186962127686
    },
    {
      "auxiliary_loss_clip": 0.01303627,
      "auxiliary_loss_mlp": 0.01046433,
      "balance_loss_clip": 1.07997203,
      "balance_loss_mlp": 1.02916002,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 2.128925001579688,
      "language_loss": 0.75119382,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77469444,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 2.6936001777648926
    },
    {
      "auxiliary_loss_clip": 0.01304256,
      "auxiliary_loss_mlp": 0.01062067,
      "balance_loss_clip": 1.08026934,
      "balance_loss_mlp": 1.04513931,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.9758965818690293,
      "language_loss": 0.82658947,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.85025269,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 2.65899658203125
    },
    {
      "auxiliary_loss_clip": 0.01301742,
      "auxiliary_loss_mlp": 0.01055802,
      "balance_loss_clip": 1.08289552,
      "balance_loss_mlp": 1.04019785,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 2.203572328725129,
      "language_loss": 0.77949649,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80307198,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 2.789612293243408
    },
    {
      "auxiliary_loss_clip": 0.01301662,
      "auxiliary_loss_mlp": 0.01047543,
      "balance_loss_clip": 1.08197379,
      "balance_loss_mlp": 1.03224826,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.042314844028655,
      "language_loss": 0.82557553,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.84906757,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.622620105743408
    },
    {
      "auxiliary_loss_clip": 0.01300818,
      "auxiliary_loss_mlp": 0.01047904,
      "balance_loss_clip": 1.07845294,
      "balance_loss_mlp": 1.03226399,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 2.7396657700666163,
      "language_loss": 0.91194439,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.9354316,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.5324411392211914
    },
    {
      "auxiliary_loss_clip": 0.01302227,
      "auxiliary_loss_mlp": 0.01055691,
      "balance_loss_clip": 1.08055401,
      "balance_loss_mlp": 1.03871596,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.6845465629639795,
      "language_loss": 0.8123908,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83596992,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.6615521907806396
    },
    {
      "auxiliary_loss_clip": 0.01298776,
      "auxiliary_loss_mlp": 0.01058912,
      "balance_loss_clip": 1.07926941,
      "balance_loss_mlp": 1.0414356,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 2.0781365609763864,
      "language_loss": 0.82619774,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.8497746,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.620232105255127
    },
    {
      "auxiliary_loss_clip": 0.01298428,
      "auxiliary_loss_mlp": 0.01056269,
      "balance_loss_clip": 1.07897282,
      "balance_loss_mlp": 1.04064035,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.5316168947577278,
      "language_loss": 0.8035959,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82714283,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.7507362365722656
    },
    {
      "auxiliary_loss_clip": 0.01299119,
      "auxiliary_loss_mlp": 0.01050803,
      "balance_loss_clip": 1.07941175,
      "balance_loss_mlp": 1.03541303,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.999163804510872,
      "language_loss": 0.8938446,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91734385,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.627148389816284
    },
    {
      "auxiliary_loss_clip": 0.01186378,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04444349,
      "balance_loss_mlp": 1.01861942,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.867349545203592,
      "language_loss": 0.63843375,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66055709,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.1799070835113525
    },
    {
      "auxiliary_loss_clip": 0.01303263,
      "auxiliary_loss_mlp": 0.01054965,
      "balance_loss_clip": 1.08323526,
      "balance_loss_mlp": 1.03859794,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 1.9924390316930016,
      "language_loss": 0.81481808,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.83840036,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.5887954235076904
    },
    {
      "auxiliary_loss_clip": 0.01297667,
      "auxiliary_loss_mlp": 0.01046297,
      "balance_loss_clip": 1.0779767,
      "balance_loss_mlp": 1.03072202,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.5974961749967584,
      "language_loss": 0.87694275,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.9003824,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 3.558976173400879
    },
    {
      "auxiliary_loss_clip": 0.01296262,
      "auxiliary_loss_mlp": 0.01056478,
      "balance_loss_clip": 1.07694197,
      "balance_loss_mlp": 1.04040885,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 2.29247423261646,
      "language_loss": 0.92591447,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.94944191,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.720494508743286
    },
    {
      "auxiliary_loss_clip": 0.01298696,
      "auxiliary_loss_mlp": 0.01058037,
      "balance_loss_clip": 1.08079696,
      "balance_loss_mlp": 1.04326677,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 2.4198906738089896,
      "language_loss": 0.77882934,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80239666,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 3.5870747566223145
    },
    {
      "auxiliary_loss_clip": 0.01302859,
      "auxiliary_loss_mlp": 0.01050786,
      "balance_loss_clip": 1.0831722,
      "balance_loss_mlp": 1.03478789,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 2.0082643791791615,
      "language_loss": 0.86994112,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89347756,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 3.5430476665496826
    },
    {
      "auxiliary_loss_clip": 0.01303239,
      "auxiliary_loss_mlp": 0.01058412,
      "balance_loss_clip": 1.08342814,
      "balance_loss_mlp": 1.04229462,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.958047529286253,
      "language_loss": 0.82854378,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.85216027,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.6246767044067383
    },
    {
      "auxiliary_loss_clip": 0.01293275,
      "auxiliary_loss_mlp": 0.01055445,
      "balance_loss_clip": 1.07644594,
      "balance_loss_mlp": 1.03831434,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.4852449458973336,
      "language_loss": 0.88627028,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90975749,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.5580341815948486
    },
    {
      "auxiliary_loss_clip": 0.0129539,
      "auxiliary_loss_mlp": 0.01052567,
      "balance_loss_clip": 1.07903719,
      "balance_loss_mlp": 1.0363065,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.6990734474999254,
      "language_loss": 0.821401,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.84488058,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.6615447998046875
    },
    {
      "auxiliary_loss_clip": 0.01292582,
      "auxiliary_loss_mlp": 0.01043253,
      "balance_loss_clip": 1.07548523,
      "balance_loss_mlp": 1.02853107,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 3.710261350954717,
      "language_loss": 0.86354399,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88690233,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.6133370399475098
    },
    {
      "auxiliary_loss_clip": 0.01294014,
      "auxiliary_loss_mlp": 0.01051221,
      "balance_loss_clip": 1.07589996,
      "balance_loss_mlp": 1.03417373,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 5.491951389752338,
      "language_loss": 0.93636489,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.95981729,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.638310432434082
    },
    {
      "auxiliary_loss_clip": 0.01193238,
      "auxiliary_loss_mlp": 0.01015823,
      "balance_loss_clip": 1.04886985,
      "balance_loss_mlp": 1.00757372,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8956094055549263,
      "language_loss": 0.63204277,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65413338,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.060819625854492
    },
    {
      "auxiliary_loss_clip": 0.01291939,
      "auxiliary_loss_mlp": 0.01053121,
      "balance_loss_clip": 1.07726359,
      "balance_loss_mlp": 1.0388751,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 1.9697749396962996,
      "language_loss": 0.79469317,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81814373,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.6634387969970703
    },
    {
      "auxiliary_loss_clip": 0.01291289,
      "auxiliary_loss_mlp": 0.0104659,
      "balance_loss_clip": 1.07511342,
      "balance_loss_mlp": 1.03027058,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 1.8418886470036455,
      "language_loss": 0.78637129,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80975008,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.6153368949890137
    },
    {
      "auxiliary_loss_clip": 0.01302406,
      "auxiliary_loss_mlp": 0.01046623,
      "balance_loss_clip": 1.07788205,
      "balance_loss_mlp": 1.03105426,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.554957705394882,
      "language_loss": 0.6167804,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64027071,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.6264588832855225
    },
    {
      "auxiliary_loss_clip": 0.01298476,
      "auxiliary_loss_mlp": 0.01047186,
      "balance_loss_clip": 1.07913172,
      "balance_loss_mlp": 1.03203499,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 3.00080889707696,
      "language_loss": 0.73661202,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76006866,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.5903139114379883
    },
    {
      "auxiliary_loss_clip": 0.01295218,
      "auxiliary_loss_mlp": 0.01051489,
      "balance_loss_clip": 1.07937932,
      "balance_loss_mlp": 1.03590798,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.9386593052493308,
      "language_loss": 0.86384749,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88731456,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 2.6777596473693848
    },
    {
      "auxiliary_loss_clip": 0.01298462,
      "auxiliary_loss_mlp": 0.01051111,
      "balance_loss_clip": 1.08033776,
      "balance_loss_mlp": 1.03532767,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 3.0188226960885505,
      "language_loss": 0.79033399,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81382966,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.7663867473602295
    },
    {
      "auxiliary_loss_clip": 0.01290407,
      "auxiliary_loss_mlp": 0.01046331,
      "balance_loss_clip": 1.07584941,
      "balance_loss_mlp": 1.03140008,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 2.035847063223271,
      "language_loss": 0.86795282,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89132017,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.5980677604675293
    },
    {
      "auxiliary_loss_clip": 0.01294456,
      "auxiliary_loss_mlp": 0.01046774,
      "balance_loss_clip": 1.07899606,
      "balance_loss_mlp": 1.03201616,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.193839707153341,
      "language_loss": 0.97558367,
      "learning_rate": 3.996069093204175e-06,
      "loss": 0.99899602,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 2.6129276752471924
    },
    {
      "auxiliary_loss_clip": 0.01300296,
      "auxiliary_loss_mlp": 0.01054342,
      "balance_loss_clip": 1.08162427,
      "balance_loss_mlp": 1.03835607,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.2188705786803427,
      "language_loss": 0.87759805,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.9011445,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.562490224838257
    },
    {
      "auxiliary_loss_clip": 0.01293562,
      "auxiliary_loss_mlp": 0.01046356,
      "balance_loss_clip": 1.07869816,
      "balance_loss_mlp": 1.03183627,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 2.207931348626015,
      "language_loss": 0.82266301,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84606218,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.637704849243164
    },
    {
      "auxiliary_loss_clip": 0.01294786,
      "auxiliary_loss_mlp": 0.01051028,
      "balance_loss_clip": 1.07700205,
      "balance_loss_mlp": 1.03562605,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.1786407138487256,
      "language_loss": 0.79577351,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.81923163,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.621272325515747
    },
    {
      "auxiliary_loss_clip": 0.01290097,
      "auxiliary_loss_mlp": 0.01046649,
      "balance_loss_clip": 1.07663846,
      "balance_loss_mlp": 1.03249288,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.994012252064582,
      "language_loss": 0.80779684,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.83116436,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.5881881713867188
    },
    {
      "auxiliary_loss_clip": 0.01196284,
      "auxiliary_loss_mlp": 0.01017124,
      "balance_loss_clip": 1.05659628,
      "balance_loss_mlp": 1.00892282,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.914776708891405,
      "language_loss": 0.59604704,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61818111,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.248509407043457
    },
    {
      "auxiliary_loss_clip": 0.01279825,
      "auxiliary_loss_mlp": 0.01050466,
      "balance_loss_clip": 1.07806087,
      "balance_loss_mlp": 1.0358274,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.138849212558437,
      "language_loss": 0.82577956,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84908241,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.7349023818969727
    },
    {
      "auxiliary_loss_clip": 0.01291799,
      "auxiliary_loss_mlp": 0.01044895,
      "balance_loss_clip": 1.07737803,
      "balance_loss_mlp": 1.02943397,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.6279646853892484,
      "language_loss": 0.97362936,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.99699634,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.5736582279205322
    },
    {
      "auxiliary_loss_clip": 0.01268214,
      "auxiliary_loss_mlp": 0.01049865,
      "balance_loss_clip": 1.07547438,
      "balance_loss_mlp": 1.03418899,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 2.712626493241647,
      "language_loss": 0.85849464,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.88167542,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.7024214267730713
    },
    {
      "auxiliary_loss_clip": 0.01293851,
      "auxiliary_loss_mlp": 0.01048498,
      "balance_loss_clip": 1.07868063,
      "balance_loss_mlp": 1.03357935,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.8072304099341276,
      "language_loss": 0.90880442,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93222785,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.616175651550293
    },
    {
      "auxiliary_loss_clip": 0.01263536,
      "auxiliary_loss_mlp": 0.01042151,
      "balance_loss_clip": 1.0700922,
      "balance_loss_mlp": 1.02694607,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 2.828603730539395,
      "language_loss": 0.80222738,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82528424,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 4.4083874225616455
    },
    {
      "auxiliary_loss_clip": 0.01262836,
      "auxiliary_loss_mlp": 0.0105245,
      "balance_loss_clip": 1.07094967,
      "balance_loss_mlp": 1.0380969,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 2.2278102486258162,
      "language_loss": 0.67472589,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.69787872,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 3.7184808254241943
    },
    {
      "auxiliary_loss_clip": 0.01266212,
      "auxiliary_loss_mlp": 0.01045451,
      "balance_loss_clip": 1.07338727,
      "balance_loss_mlp": 1.03064501,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 2.0911426343460615,
      "language_loss": 0.76789463,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.79101133,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.7602291107177734
    },
    {
      "auxiliary_loss_clip": 0.012682,
      "auxiliary_loss_mlp": 0.01052584,
      "balance_loss_clip": 1.07225728,
      "balance_loss_mlp": 1.03762364,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.0698784969030926,
      "language_loss": 0.8073656,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.83057344,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 3.632122039794922
    },
    {
      "auxiliary_loss_clip": 0.01268046,
      "auxiliary_loss_mlp": 0.01051621,
      "balance_loss_clip": 1.07431281,
      "balance_loss_mlp": 1.03629637,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.3667663937229544,
      "language_loss": 0.9331699,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95636666,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.7008159160614014
    },
    {
      "auxiliary_loss_clip": 0.01252443,
      "auxiliary_loss_mlp": 0.0090472,
      "balance_loss_clip": 1.0714463,
      "balance_loss_mlp": 1.00052357,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.7557885456068025,
      "language_loss": 0.75619835,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77777004,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.751349925994873
    },
    {
      "auxiliary_loss_clip": 0.01258739,
      "auxiliary_loss_mlp": 0.01052468,
      "balance_loss_clip": 1.0709331,
      "balance_loss_mlp": 1.0357554,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 2.6748417390055175,
      "language_loss": 0.83280861,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85592067,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.7253541946411133
    },
    {
      "auxiliary_loss_clip": 0.01275229,
      "auxiliary_loss_mlp": 0.01043897,
      "balance_loss_clip": 1.07472575,
      "balance_loss_mlp": 1.0291748,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 2.52965687209236,
      "language_loss": 0.85099018,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87418145,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.7625255584716797
    },
    {
      "auxiliary_loss_clip": 0.01289062,
      "auxiliary_loss_mlp": 0.01050435,
      "balance_loss_clip": 1.07668006,
      "balance_loss_mlp": 1.03549826,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 1.7900690002731554,
      "language_loss": 0.85602766,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87942261,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.648467779159546
    },
    {
      "auxiliary_loss_clip": 0.01171887,
      "auxiliary_loss_mlp": 0.00896269,
      "balance_loss_clip": 1.04967058,
      "balance_loss_mlp": 1.00025463,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8814472341126928,
      "language_loss": 0.63097602,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.65165752,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.249635934829712
    },
    {
      "auxiliary_loss_clip": 0.0119646,
      "auxiliary_loss_mlp": 0.00896234,
      "balance_loss_clip": 1.05268502,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8595149336069339,
      "language_loss": 0.59944856,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.62037551,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.1756386756896973
    },
    {
      "auxiliary_loss_clip": 0.01179558,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.04521024,
      "balance_loss_mlp": 1.01814878,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9429606057179932,
      "language_loss": 0.6314311,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65348256,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.9107449054718018
    },
    {
      "auxiliary_loss_clip": 0.01262436,
      "auxiliary_loss_mlp": 0.01050968,
      "balance_loss_clip": 1.06972277,
      "balance_loss_mlp": 1.03564978,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 2.0100342203891737,
      "language_loss": 0.79451406,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81764811,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.623960256576538
    },
    {
      "auxiliary_loss_clip": 0.01188716,
      "auxiliary_loss_mlp": 0.01019665,
      "balance_loss_clip": 1.04714191,
      "balance_loss_mlp": 1.01232135,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8419838558713707,
      "language_loss": 0.5529536,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57503748,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.0935401916503906
    },
    {
      "auxiliary_loss_clip": 0.01223319,
      "auxiliary_loss_mlp": 0.01054051,
      "balance_loss_clip": 1.05775976,
      "balance_loss_mlp": 1.03839922,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.512728661453403,
      "language_loss": 0.78728187,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.81005555,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 2.7635815143585205
    },
    {
      "auxiliary_loss_clip": 0.01238711,
      "auxiliary_loss_mlp": 0.01056837,
      "balance_loss_clip": 1.06573629,
      "balance_loss_mlp": 1.04283595,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 2.4837272084136917,
      "language_loss": 0.85935879,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88231421,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.7417449951171875
    },
    {
      "auxiliary_loss_clip": 0.01222901,
      "auxiliary_loss_mlp": 0.01044137,
      "balance_loss_clip": 1.06480312,
      "balance_loss_mlp": 1.02986813,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 2.2355839054890496,
      "language_loss": 0.9351086,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95777893,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 3.018190622329712
    },
    {
      "auxiliary_loss_clip": 0.0126479,
      "auxiliary_loss_mlp": 0.01055165,
      "balance_loss_clip": 1.07521391,
      "balance_loss_mlp": 1.04000115,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.9197881277268756,
      "language_loss": 0.91392839,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93712795,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 2.9716997146606445
    },
    {
      "auxiliary_loss_clip": 0.01272965,
      "auxiliary_loss_mlp": 0.01039934,
      "balance_loss_clip": 1.07400703,
      "balance_loss_mlp": 1.02587867,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 2.027261741344619,
      "language_loss": 0.81228769,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83541667,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.7068262100219727
    },
    {
      "auxiliary_loss_clip": 0.0128651,
      "auxiliary_loss_mlp": 0.01056748,
      "balance_loss_clip": 1.07862854,
      "balance_loss_mlp": 1.0424304,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.2432915067286627,
      "language_loss": 0.92666793,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95010048,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.6589698791503906
    },
    {
      "auxiliary_loss_clip": 0.01169313,
      "auxiliary_loss_mlp": 0.01013009,
      "balance_loss_clip": 1.03988886,
      "balance_loss_mlp": 1.00609493,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8933189634295677,
      "language_loss": 0.59265476,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61447799,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.3025972843170166
    },
    {
      "auxiliary_loss_clip": 0.01238463,
      "auxiliary_loss_mlp": 0.01050118,
      "balance_loss_clip": 1.06530571,
      "balance_loss_mlp": 1.03572989,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.397137244553029,
      "language_loss": 0.92424214,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94712794,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.6962015628814697
    },
    {
      "auxiliary_loss_clip": 0.01271826,
      "auxiliary_loss_mlp": 0.01047985,
      "balance_loss_clip": 1.07186162,
      "balance_loss_mlp": 1.03322649,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.4724900804656733,
      "language_loss": 0.82960606,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85280418,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.6197259426116943
    },
    {
      "auxiliary_loss_clip": 0.01274903,
      "auxiliary_loss_mlp": 0.01049937,
      "balance_loss_clip": 1.07522893,
      "balance_loss_mlp": 1.03536975,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 2.264755133390913,
      "language_loss": 0.73995423,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76320267,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.736074686050415
    },
    {
      "auxiliary_loss_clip": 0.01253777,
      "auxiliary_loss_mlp": 0.01046289,
      "balance_loss_clip": 1.07176161,
      "balance_loss_mlp": 1.03019536,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.712138271053934,
      "language_loss": 0.84933341,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87233406,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.69889235496521
    },
    {
      "auxiliary_loss_clip": 0.01264045,
      "auxiliary_loss_mlp": 0.01052037,
      "balance_loss_clip": 1.07653022,
      "balance_loss_mlp": 1.03656328,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 2.121656259323995,
      "language_loss": 0.81808507,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.84124601,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.726907968521118
    },
    {
      "auxiliary_loss_clip": 0.01273859,
      "auxiliary_loss_mlp": 0.01040814,
      "balance_loss_clip": 1.07219732,
      "balance_loss_mlp": 1.02731323,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.0791195109632254,
      "language_loss": 0.78584951,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80899632,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 3.6258621215820312
    },
    {
      "auxiliary_loss_clip": 0.01285835,
      "auxiliary_loss_mlp": 0.01049484,
      "balance_loss_clip": 1.07657039,
      "balance_loss_mlp": 1.03529215,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 2.023015863371551,
      "language_loss": 0.82340896,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84676218,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 4.4747474193573
    },
    {
      "auxiliary_loss_clip": 0.0127514,
      "auxiliary_loss_mlp": 0.01056257,
      "balance_loss_clip": 1.07689595,
      "balance_loss_mlp": 1.04227948,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 2.333782717644816,
      "language_loss": 0.868303,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89161694,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.655062675476074
    },
    {
      "auxiliary_loss_clip": 0.01243036,
      "auxiliary_loss_mlp": 0.01044175,
      "balance_loss_clip": 1.06437945,
      "balance_loss_mlp": 1.02983975,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.6365852998499442,
      "language_loss": 0.77212608,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79499817,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 3.6632118225097656
    },
    {
      "auxiliary_loss_clip": 0.0127955,
      "auxiliary_loss_mlp": 0.01049481,
      "balance_loss_clip": 1.07606149,
      "balance_loss_mlp": 1.03572464,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 1.9689912938405532,
      "language_loss": 0.73717833,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76046866,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.6414740085601807
    },
    {
      "auxiliary_loss_clip": 0.01273195,
      "auxiliary_loss_mlp": 0.01051299,
      "balance_loss_clip": 1.07400846,
      "balance_loss_mlp": 1.03667188,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 2.608745777206029,
      "language_loss": 0.86498648,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.8882314,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.6362645626068115
    },
    {
      "auxiliary_loss_clip": 0.01273276,
      "auxiliary_loss_mlp": 0.01046226,
      "balance_loss_clip": 1.07603538,
      "balance_loss_mlp": 1.03176618,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.3649353628782026,
      "language_loss": 0.92787045,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95106554,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.6607248783111572
    },
    {
      "auxiliary_loss_clip": 0.01257666,
      "auxiliary_loss_mlp": 0.01049938,
      "balance_loss_clip": 1.07388997,
      "balance_loss_mlp": 1.0351088,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.484407167599684,
      "language_loss": 0.87433606,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89741212,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.6289052963256836
    },
    {
      "auxiliary_loss_clip": 0.01265557,
      "auxiliary_loss_mlp": 0.01050153,
      "balance_loss_clip": 1.07288158,
      "balance_loss_mlp": 1.03592527,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 3.4706488215784312,
      "language_loss": 0.87725621,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.90041333,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.6209018230438232
    },
    {
      "auxiliary_loss_clip": 0.01260258,
      "auxiliary_loss_mlp": 0.01049044,
      "balance_loss_clip": 1.07243967,
      "balance_loss_mlp": 1.03528142,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.9786911189156524,
      "language_loss": 0.71202886,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73512197,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.919794797897339
    },
    {
      "auxiliary_loss_clip": 0.01252402,
      "auxiliary_loss_mlp": 0.01050733,
      "balance_loss_clip": 1.07153535,
      "balance_loss_mlp": 1.03644633,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 1.9365449523249498,
      "language_loss": 0.83873153,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86176294,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.648197889328003
    },
    {
      "auxiliary_loss_clip": 0.01211887,
      "auxiliary_loss_mlp": 0.01044868,
      "balance_loss_clip": 1.05997252,
      "balance_loss_mlp": 1.02999055,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.7183192740277413,
      "language_loss": 0.81037569,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.8329432,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 3.1136558055877686
    },
    {
      "auxiliary_loss_clip": 0.01240435,
      "auxiliary_loss_mlp": 0.01051421,
      "balance_loss_clip": 1.06641877,
      "balance_loss_mlp": 1.03641295,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 4.944453862317652,
      "language_loss": 0.93265617,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95557475,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 3.0822978019714355
    },
    {
      "auxiliary_loss_clip": 0.01226667,
      "auxiliary_loss_mlp": 0.01051101,
      "balance_loss_clip": 1.06504261,
      "balance_loss_mlp": 1.03667641,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.6695138311629363,
      "language_loss": 0.86278886,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88556647,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.8465359210968018
    },
    {
      "auxiliary_loss_clip": 0.01264266,
      "auxiliary_loss_mlp": 0.01043564,
      "balance_loss_clip": 1.06730103,
      "balance_loss_mlp": 1.02984858,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 2.184480502348525,
      "language_loss": 0.79886007,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.8219384,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 2.719498872756958
    },
    {
      "auxiliary_loss_clip": 0.01281425,
      "auxiliary_loss_mlp": 0.01046959,
      "balance_loss_clip": 1.07280529,
      "balance_loss_mlp": 1.03319681,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 2.3087189766648573,
      "language_loss": 0.87034875,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89363265,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.567028760910034
    },
    {
      "auxiliary_loss_clip": 0.01256477,
      "auxiliary_loss_mlp": 0.01041579,
      "balance_loss_clip": 1.0697279,
      "balance_loss_mlp": 1.02794719,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.0513905585171823,
      "language_loss": 0.82390857,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84688908,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 2.7163314819335938
    },
    {
      "auxiliary_loss_clip": 0.01261099,
      "auxiliary_loss_mlp": 0.01043435,
      "balance_loss_clip": 1.0669167,
      "balance_loss_mlp": 1.02953529,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.586161934908489,
      "language_loss": 0.80112505,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82417041,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.7325539588928223
    },
    {
      "auxiliary_loss_clip": 0.01232031,
      "auxiliary_loss_mlp": 0.01044354,
      "balance_loss_clip": 1.0661993,
      "balance_loss_mlp": 1.02952445,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 1.969103331643808,
      "language_loss": 0.84428686,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86705071,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.8086297512054443
    },
    {
      "auxiliary_loss_clip": 0.01280074,
      "auxiliary_loss_mlp": 0.01047187,
      "balance_loss_clip": 1.07449257,
      "balance_loss_mlp": 1.03260756,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.9035046772044306,
      "language_loss": 0.89630556,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.91957819,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.6146013736724854
    },
    {
      "auxiliary_loss_clip": 0.01272143,
      "auxiliary_loss_mlp": 0.01049543,
      "balance_loss_clip": 1.07030296,
      "balance_loss_mlp": 1.03560758,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 6.070878666808316,
      "language_loss": 0.88274926,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.9059661,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.6229419708251953
    },
    {
      "auxiliary_loss_clip": 0.01236223,
      "auxiliary_loss_mlp": 0.01041591,
      "balance_loss_clip": 1.06698298,
      "balance_loss_mlp": 1.02673733,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.047326858260539,
      "language_loss": 0.81064785,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.833426,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.7276973724365234
    },
    {
      "auxiliary_loss_clip": 0.01250249,
      "auxiliary_loss_mlp": 0.0105486,
      "balance_loss_clip": 1.07080996,
      "balance_loss_mlp": 1.04101372,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.2480265646499884,
      "language_loss": 0.8516748,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87472594,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.7036499977111816
    },
    {
      "auxiliary_loss_clip": 0.01256215,
      "auxiliary_loss_mlp": 0.01044985,
      "balance_loss_clip": 1.06855345,
      "balance_loss_mlp": 1.03088808,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 2.43215721536085,
      "language_loss": 0.8083843,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.83139622,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.680138349533081
    },
    {
      "auxiliary_loss_clip": 0.01246623,
      "auxiliary_loss_mlp": 0.01044832,
      "balance_loss_clip": 1.06865239,
      "balance_loss_mlp": 1.0300622,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.0856329910086377,
      "language_loss": 0.88153827,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90445286,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.6545958518981934
    },
    {
      "auxiliary_loss_clip": 0.01235023,
      "auxiliary_loss_mlp": 0.01047296,
      "balance_loss_clip": 1.06523967,
      "balance_loss_mlp": 1.03340244,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 1.914677133687072,
      "language_loss": 0.86069119,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.8835144,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.7212653160095215
    },
    {
      "auxiliary_loss_clip": 0.01280247,
      "auxiliary_loss_mlp": 0.00904854,
      "balance_loss_clip": 1.07712412,
      "balance_loss_mlp": 1.00035834,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.314236753811737,
      "language_loss": 0.82751739,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.84936833,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.8571207523345947
    },
    {
      "auxiliary_loss_clip": 0.01270173,
      "auxiliary_loss_mlp": 0.01048457,
      "balance_loss_clip": 1.0740037,
      "balance_loss_mlp": 1.03422964,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.421443398161905,
      "language_loss": 0.83990186,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86308813,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 4.610533237457275
    },
    {
      "auxiliary_loss_clip": 0.01283715,
      "auxiliary_loss_mlp": 0.01054592,
      "balance_loss_clip": 1.07669497,
      "balance_loss_mlp": 1.03956008,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.43747924107555,
      "language_loss": 0.80532682,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.8287099,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 3.571441888809204
    },
    {
      "auxiliary_loss_clip": 0.01234309,
      "auxiliary_loss_mlp": 0.01045489,
      "balance_loss_clip": 1.06697822,
      "balance_loss_mlp": 1.0316484,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.4073746080876077,
      "language_loss": 0.86495131,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88774931,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.7140769958496094
    },
    {
      "auxiliary_loss_clip": 0.01281934,
      "auxiliary_loss_mlp": 0.0104852,
      "balance_loss_clip": 1.07769251,
      "balance_loss_mlp": 1.03436971,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.2334697256093814,
      "language_loss": 0.76602745,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.78933191,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 3.7264678478240967
    },
    {
      "auxiliary_loss_clip": 0.01257568,
      "auxiliary_loss_mlp": 0.01052259,
      "balance_loss_clip": 1.07344687,
      "balance_loss_mlp": 1.03754282,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.0530780443245042,
      "language_loss": 0.90543711,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92853534,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.694737672805786
    },
    {
      "auxiliary_loss_clip": 0.01248377,
      "auxiliary_loss_mlp": 0.00904706,
      "balance_loss_clip": 1.06648469,
      "balance_loss_mlp": 1.0003686,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 2.044307983935373,
      "language_loss": 0.82149845,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84302932,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.7853798866271973
    },
    {
      "auxiliary_loss_clip": 0.01159378,
      "auxiliary_loss_mlp": 0.01002535,
      "balance_loss_clip": 1.04244864,
      "balance_loss_mlp": 0.99664634,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.8441839339372709,
      "language_loss": 0.55034494,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57196403,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.2547826766967773
    },
    {
      "auxiliary_loss_clip": 0.0124375,
      "auxiliary_loss_mlp": 0.01050865,
      "balance_loss_clip": 1.06323647,
      "balance_loss_mlp": 1.0362258,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 2.1601748883892555,
      "language_loss": 0.78379059,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.80673683,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.7118003368377686
    },
    {
      "auxiliary_loss_clip": 0.01280904,
      "auxiliary_loss_mlp": 0.01042484,
      "balance_loss_clip": 1.07718706,
      "balance_loss_mlp": 1.02816713,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 1.9577410949874905,
      "language_loss": 0.80753529,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83076918,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.8594863414764404
    },
    {
      "auxiliary_loss_clip": 0.01277021,
      "auxiliary_loss_mlp": 0.01048086,
      "balance_loss_clip": 1.07638621,
      "balance_loss_mlp": 1.03399515,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.2440861374156205,
      "language_loss": 0.77654052,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79979163,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.64510440826416
    },
    {
      "auxiliary_loss_clip": 0.0125633,
      "auxiliary_loss_mlp": 0.01041806,
      "balance_loss_clip": 1.06778884,
      "balance_loss_mlp": 1.02881837,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.407211877660254,
      "language_loss": 0.87715733,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.90013874,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.616025924682617
    },
    {
      "auxiliary_loss_clip": 0.01270199,
      "auxiliary_loss_mlp": 0.01052077,
      "balance_loss_clip": 1.07355797,
      "balance_loss_mlp": 1.03835607,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 2.144191352070656,
      "language_loss": 0.78079712,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80401993,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.9895355701446533
    },
    {
      "auxiliary_loss_clip": 0.01239477,
      "auxiliary_loss_mlp": 0.01045093,
      "balance_loss_clip": 1.06605053,
      "balance_loss_mlp": 1.03090715,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 4.298281554495898,
      "language_loss": 0.87680852,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.89965427,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.6890084743499756
    },
    {
      "auxiliary_loss_clip": 0.01263468,
      "auxiliary_loss_mlp": 0.01050665,
      "balance_loss_clip": 1.07190764,
      "balance_loss_mlp": 1.03668725,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 1.8595972944359334,
      "language_loss": 0.87839401,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90153527,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 2.7070868015289307
    },
    {
      "auxiliary_loss_clip": 0.01271894,
      "auxiliary_loss_mlp": 0.01047035,
      "balance_loss_clip": 1.07671118,
      "balance_loss_mlp": 1.03239608,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 4.158743732992043,
      "language_loss": 0.80623353,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.82942283,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 2.6916873455047607
    },
    {
      "auxiliary_loss_clip": 0.01245115,
      "auxiliary_loss_mlp": 0.01046812,
      "balance_loss_clip": 1.069502,
      "balance_loss_mlp": 1.03374624,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.4205207917832654,
      "language_loss": 0.78020906,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.8031283,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 2.7436347007751465
    },
    {
      "auxiliary_loss_clip": 0.01267657,
      "auxiliary_loss_mlp": 0.00905436,
      "balance_loss_clip": 1.07267737,
      "balance_loss_mlp": 1.00046217,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.1941787489226745,
      "language_loss": 0.81692052,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.83865148,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.6692771911621094
    },
    {
      "auxiliary_loss_clip": 0.0125928,
      "auxiliary_loss_mlp": 0.0105029,
      "balance_loss_clip": 1.0696981,
      "balance_loss_mlp": 1.03395855,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 2.7113859813308516,
      "language_loss": 0.87948447,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90258014,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.725403070449829
    },
    {
      "auxiliary_loss_clip": 0.01280925,
      "auxiliary_loss_mlp": 0.01045089,
      "balance_loss_clip": 1.07861507,
      "balance_loss_mlp": 1.03074789,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.17207726974377,
      "language_loss": 0.81458288,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.83784294,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.6518514156341553
    },
    {
      "auxiliary_loss_clip": 0.01285202,
      "auxiliary_loss_mlp": 0.01050378,
      "balance_loss_clip": 1.07578421,
      "balance_loss_mlp": 1.0366689,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 3.189697267677509,
      "language_loss": 0.80721921,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83057499,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.59605073928833
    },
    {
      "auxiliary_loss_clip": 0.01267125,
      "auxiliary_loss_mlp": 0.0104377,
      "balance_loss_clip": 1.07035279,
      "balance_loss_mlp": 1.02973938,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.241037467562022,
      "language_loss": 0.76153553,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78464448,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.719870090484619
    },
    {
      "auxiliary_loss_clip": 0.01266393,
      "auxiliary_loss_mlp": 0.01047774,
      "balance_loss_clip": 1.07395065,
      "balance_loss_mlp": 1.03371918,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.0033460304758752,
      "language_loss": 0.78367889,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.80682063,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.6580002307891846
    },
    {
      "auxiliary_loss_clip": 0.01240313,
      "auxiliary_loss_mlp": 0.00906046,
      "balance_loss_clip": 1.06417227,
      "balance_loss_mlp": 1.00036812,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 1.9738465550543765,
      "language_loss": 0.90836608,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92982966,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.7272543907165527
    },
    {
      "auxiliary_loss_clip": 0.01256239,
      "auxiliary_loss_mlp": 0.01056292,
      "balance_loss_clip": 1.06850278,
      "balance_loss_mlp": 1.04146254,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.1851019138130736,
      "language_loss": 0.93164605,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.9547714,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.678447723388672
    },
    {
      "auxiliary_loss_clip": 0.01281259,
      "auxiliary_loss_mlp": 0.01045006,
      "balance_loss_clip": 1.07677746,
      "balance_loss_mlp": 1.0306232,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.794472116326211,
      "language_loss": 0.7786231,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80188572,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.6817803382873535
    },
    {
      "auxiliary_loss_clip": 0.01246156,
      "auxiliary_loss_mlp": 0.01055067,
      "balance_loss_clip": 1.07330179,
      "balance_loss_mlp": 1.03971291,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.2498486041941996,
      "language_loss": 0.81309724,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.8361094,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.698042869567871
    },
    {
      "auxiliary_loss_clip": 0.01237478,
      "auxiliary_loss_mlp": 0.01049051,
      "balance_loss_clip": 1.06982362,
      "balance_loss_mlp": 1.03449559,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.250370814397565,
      "language_loss": 0.84702075,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.86988598,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.739689350128174
    },
    {
      "auxiliary_loss_clip": 0.01255094,
      "auxiliary_loss_mlp": 0.00904739,
      "balance_loss_clip": 1.07082248,
      "balance_loss_mlp": 1.00038326,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.9158031990121107,
      "language_loss": 0.91083384,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.93243223,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 3.7171125411987305
    },
    {
      "auxiliary_loss_clip": 0.01246646,
      "auxiliary_loss_mlp": 0.01055482,
      "balance_loss_clip": 1.06723475,
      "balance_loss_mlp": 1.04052162,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.193111142898653,
      "language_loss": 0.93764716,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96066844,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 3.6229758262634277
    },
    {
      "auxiliary_loss_clip": 0.01256811,
      "auxiliary_loss_mlp": 0.0105293,
      "balance_loss_clip": 1.07294166,
      "balance_loss_mlp": 1.03836226,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 1.881920569883086,
      "language_loss": 0.71310306,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73620045,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 3.587087869644165
    },
    {
      "auxiliary_loss_clip": 0.01266649,
      "auxiliary_loss_mlp": 0.00905263,
      "balance_loss_clip": 1.07324219,
      "balance_loss_mlp": 1.00032163,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 1.8478624467541764,
      "language_loss": 0.7840482,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.8057673,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.769071102142334
    },
    {
      "auxiliary_loss_clip": 0.01246526,
      "auxiliary_loss_mlp": 0.01047242,
      "balance_loss_clip": 1.0678761,
      "balance_loss_mlp": 1.03343105,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.775472394010771,
      "language_loss": 0.74294209,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76587987,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.635230779647827
    },
    {
      "auxiliary_loss_clip": 0.0125986,
      "auxiliary_loss_mlp": 0.01045913,
      "balance_loss_clip": 1.07092333,
      "balance_loss_mlp": 1.0326215,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 2.46304263380414,
      "language_loss": 0.91928113,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94233888,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.6972720623016357
    },
    {
      "auxiliary_loss_clip": 0.01251714,
      "auxiliary_loss_mlp": 0.01050266,
      "balance_loss_clip": 1.06528497,
      "balance_loss_mlp": 1.0354718,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.7776030932914564,
      "language_loss": 0.6753872,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.69840693,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.651874303817749
    },
    {
      "auxiliary_loss_clip": 0.01227725,
      "auxiliary_loss_mlp": 0.01053745,
      "balance_loss_clip": 1.06391275,
      "balance_loss_mlp": 1.03928483,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.743784895975508,
      "language_loss": 0.94859469,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97140932,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.687103509902954
    },
    {
      "auxiliary_loss_clip": 0.01241817,
      "auxiliary_loss_mlp": 0.01048536,
      "balance_loss_clip": 1.06606805,
      "balance_loss_mlp": 1.03498173,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.3356115903661756,
      "language_loss": 0.75234878,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77525234,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.8009495735168457
    },
    {
      "auxiliary_loss_clip": 0.01269112,
      "auxiliary_loss_mlp": 0.01051448,
      "balance_loss_clip": 1.0739913,
      "balance_loss_mlp": 1.03672624,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 2.4804214614597817,
      "language_loss": 0.81162524,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83483082,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.717376708984375
    },
    {
      "auxiliary_loss_clip": 0.01264096,
      "auxiliary_loss_mlp": 0.01043565,
      "balance_loss_clip": 1.07683885,
      "balance_loss_mlp": 1.02976108,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 2.0089800426784343,
      "language_loss": 0.79983866,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82291532,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.6503312587738037
    },
    {
      "auxiliary_loss_clip": 0.0126673,
      "auxiliary_loss_mlp": 0.01040008,
      "balance_loss_clip": 1.07554424,
      "balance_loss_mlp": 1.02605486,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.0465825468128656,
      "language_loss": 0.94077611,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96384346,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.681717872619629
    },
    {
      "auxiliary_loss_clip": 0.01216776,
      "auxiliary_loss_mlp": 0.01045435,
      "balance_loss_clip": 1.06836152,
      "balance_loss_mlp": 1.03101063,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.94743485144785,
      "language_loss": 0.82960451,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85222661,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.6688907146453857
    },
    {
      "auxiliary_loss_clip": 0.01202186,
      "auxiliary_loss_mlp": 0.01045394,
      "balance_loss_clip": 1.0574764,
      "balance_loss_mlp": 1.03083849,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 3.005223596382007,
      "language_loss": 0.80253744,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82501328,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 2.7901625633239746
    },
    {
      "auxiliary_loss_clip": 0.01267339,
      "auxiliary_loss_mlp": 0.01044802,
      "balance_loss_clip": 1.07055461,
      "balance_loss_mlp": 1.03052068,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.620112222643464,
      "language_loss": 0.81004703,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83316845,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.621692180633545
    },
    {
      "auxiliary_loss_clip": 0.01277961,
      "auxiliary_loss_mlp": 0.01049705,
      "balance_loss_clip": 1.07504272,
      "balance_loss_mlp": 1.03518534,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.166342451195799,
      "language_loss": 0.86158335,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88486004,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.6390461921691895
    },
    {
      "auxiliary_loss_clip": 0.0124806,
      "auxiliary_loss_mlp": 0.01049959,
      "balance_loss_clip": 1.06670582,
      "balance_loss_mlp": 1.03461647,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.2315583253903792,
      "language_loss": 0.76804948,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79102963,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 2.753183126449585
    },
    {
      "auxiliary_loss_clip": 0.01252914,
      "auxiliary_loss_mlp": 0.01050001,
      "balance_loss_clip": 1.07211709,
      "balance_loss_mlp": 1.03494465,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.039402238638009,
      "language_loss": 0.84476793,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86779702,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.6925134658813477
    },
    {
      "auxiliary_loss_clip": 0.01265399,
      "auxiliary_loss_mlp": 0.01041511,
      "balance_loss_clip": 1.06719184,
      "balance_loss_mlp": 1.02760553,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.1709431966055788,
      "language_loss": 0.94459915,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.96766829,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.6365926265716553
    },
    {
      "auxiliary_loss_clip": 0.01247147,
      "auxiliary_loss_mlp": 0.01047069,
      "balance_loss_clip": 1.0684886,
      "balance_loss_mlp": 1.03281116,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.2972050681480987,
      "language_loss": 0.86515397,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.88809615,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.653576135635376
    },
    {
      "auxiliary_loss_clip": 0.01247952,
      "auxiliary_loss_mlp": 0.01046189,
      "balance_loss_clip": 1.06901598,
      "balance_loss_mlp": 1.03106189,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 3.4131024109874057,
      "language_loss": 0.73722529,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.76016665,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.8040149211883545
    },
    {
      "auxiliary_loss_clip": 0.0125695,
      "auxiliary_loss_mlp": 0.0104477,
      "balance_loss_clip": 1.07181799,
      "balance_loss_mlp": 1.0301187,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.2993559867106224,
      "language_loss": 0.79080069,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81381786,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.7246310710906982
    },
    {
      "auxiliary_loss_clip": 0.01275513,
      "auxiliary_loss_mlp": 0.0105041,
      "balance_loss_clip": 1.0744307,
      "balance_loss_mlp": 1.03644443,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.5764203025345314,
      "language_loss": 0.87541425,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89867342,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.7742512226104736
    },
    {
      "auxiliary_loss_clip": 0.01263264,
      "auxiliary_loss_mlp": 0.01052225,
      "balance_loss_clip": 1.07467127,
      "balance_loss_mlp": 1.03889167,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.702196978743646,
      "language_loss": 0.79947239,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82262731,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.707432270050049
    },
    {
      "auxiliary_loss_clip": 0.01270872,
      "auxiliary_loss_mlp": 0.01045936,
      "balance_loss_clip": 1.07405138,
      "balance_loss_mlp": 1.03252494,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 2.367330331268858,
      "language_loss": 0.91890746,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.94207555,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.6419527530670166
    },
    {
      "auxiliary_loss_clip": 0.01257414,
      "auxiliary_loss_mlp": 0.01045927,
      "balance_loss_clip": 1.07139146,
      "balance_loss_mlp": 1.0333147,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.21958094898978,
      "language_loss": 0.88148558,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90451902,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.6986870765686035
    },
    {
      "auxiliary_loss_clip": 0.01257457,
      "auxiliary_loss_mlp": 0.01047387,
      "balance_loss_clip": 1.07144058,
      "balance_loss_mlp": 1.03354073,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.16939905808208,
      "language_loss": 0.83897424,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86202276,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 3.564781427383423
    },
    {
      "auxiliary_loss_clip": 0.01264043,
      "auxiliary_loss_mlp": 0.01043858,
      "balance_loss_clip": 1.07458591,
      "balance_loss_mlp": 1.02988088,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.3188435411547053,
      "language_loss": 0.76284587,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78592491,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 4.5013344287872314
    },
    {
      "auxiliary_loss_clip": 0.01260611,
      "auxiliary_loss_mlp": 0.01044882,
      "balance_loss_clip": 1.07467604,
      "balance_loss_mlp": 1.0319891,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.0186694489286663,
      "language_loss": 0.90260351,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.9256584,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.6406309604644775
    },
    {
      "auxiliary_loss_clip": 0.01238609,
      "auxiliary_loss_mlp": 0.01045367,
      "balance_loss_clip": 1.06420183,
      "balance_loss_mlp": 1.03166378,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.874832882115848,
      "language_loss": 0.8528316,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87567139,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 3.658364772796631
    },
    {
      "auxiliary_loss_clip": 0.0123365,
      "auxiliary_loss_mlp": 0.0090548,
      "balance_loss_clip": 1.06837881,
      "balance_loss_mlp": 1.00030112,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.9650076395978373,
      "language_loss": 0.89106542,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91245675,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.7052624225616455
    },
    {
      "auxiliary_loss_clip": 0.01264258,
      "auxiliary_loss_mlp": 0.01050694,
      "balance_loss_clip": 1.07300389,
      "balance_loss_mlp": 1.03725958,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 3.5318761543805715,
      "language_loss": 0.88301998,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.90616953,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.691938638687134
    },
    {
      "auxiliary_loss_clip": 0.01252548,
      "auxiliary_loss_mlp": 0.01053116,
      "balance_loss_clip": 1.06778908,
      "balance_loss_mlp": 1.0399909,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.172558997154164,
      "language_loss": 0.76447582,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.78753251,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.605510950088501
    },
    {
      "auxiliary_loss_clip": 0.01239334,
      "auxiliary_loss_mlp": 0.01044215,
      "balance_loss_clip": 1.06688261,
      "balance_loss_mlp": 1.03031468,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 1.9840656866795765,
      "language_loss": 0.76892251,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79175794,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.7921793460845947
    },
    {
      "auxiliary_loss_clip": 0.01274521,
      "auxiliary_loss_mlp": 0.01037775,
      "balance_loss_clip": 1.07538795,
      "balance_loss_mlp": 1.02437615,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 1.9573700543744534,
      "language_loss": 0.81019825,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83332121,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.6595613956451416
    },
    {
      "auxiliary_loss_clip": 0.01255265,
      "auxiliary_loss_mlp": 0.01038612,
      "balance_loss_clip": 1.07045388,
      "balance_loss_mlp": 1.02459931,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.5551551426817616,
      "language_loss": 0.85171795,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87465668,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.67718505859375
    },
    {
      "auxiliary_loss_clip": 0.01257078,
      "auxiliary_loss_mlp": 0.01041297,
      "balance_loss_clip": 1.06971526,
      "balance_loss_mlp": 1.02825594,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.3615147444277684,
      "language_loss": 0.77339828,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79638207,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.586310625076294
    },
    {
      "auxiliary_loss_clip": 0.01248986,
      "auxiliary_loss_mlp": 0.01052018,
      "balance_loss_clip": 1.07146704,
      "balance_loss_mlp": 1.0378561,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 2.6288030297941676,
      "language_loss": 0.9594183,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98242837,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.644545555114746
    },
    {
      "auxiliary_loss_clip": 0.01235444,
      "auxiliary_loss_mlp": 0.01045484,
      "balance_loss_clip": 1.06735349,
      "balance_loss_mlp": 1.0320487,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 1.9461722561740915,
      "language_loss": 0.88104391,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90385318,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.7252542972564697
    },
    {
      "auxiliary_loss_clip": 0.01249234,
      "auxiliary_loss_mlp": 0.01043721,
      "balance_loss_clip": 1.07070327,
      "balance_loss_mlp": 1.03044713,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.7998369793600217,
      "language_loss": 0.82651353,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84944308,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.73530912399292
    },
    {
      "auxiliary_loss_clip": 0.01232054,
      "auxiliary_loss_mlp": 0.01045022,
      "balance_loss_clip": 1.06355548,
      "balance_loss_mlp": 1.03146827,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.729111436301274,
      "language_loss": 0.70025432,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72302514,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.6775496006011963
    },
    {
      "auxiliary_loss_clip": 0.01261721,
      "auxiliary_loss_mlp": 0.00904348,
      "balance_loss_clip": 1.07134628,
      "balance_loss_mlp": 1.00039637,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.2298462850218836,
      "language_loss": 0.9523114,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.97397208,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 2.6340060234069824
    },
    {
      "auxiliary_loss_clip": 0.01235534,
      "auxiliary_loss_mlp": 0.0104096,
      "balance_loss_clip": 1.06927979,
      "balance_loss_mlp": 1.02790105,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 3.0236099300905965,
      "language_loss": 0.91042882,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.9331938,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.669848680496216
    },
    {
      "auxiliary_loss_clip": 0.01260547,
      "auxiliary_loss_mlp": 0.01044719,
      "balance_loss_clip": 1.070207,
      "balance_loss_mlp": 1.03133726,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.0673158589631795,
      "language_loss": 0.93150651,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95455921,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.6719183921813965
    },
    {
      "auxiliary_loss_clip": 0.01250918,
      "auxiliary_loss_mlp": 0.01051473,
      "balance_loss_clip": 1.06920719,
      "balance_loss_mlp": 1.03679836,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 3.8613120843371767,
      "language_loss": 0.8015908,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82461476,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.676938533782959
    },
    {
      "auxiliary_loss_clip": 0.01249867,
      "auxiliary_loss_mlp": 0.0105163,
      "balance_loss_clip": 1.06677186,
      "balance_loss_mlp": 1.03711605,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.0038620327990144,
      "language_loss": 0.81470406,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83771902,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.6547598838806152
    },
    {
      "auxiliary_loss_clip": 0.01255913,
      "auxiliary_loss_mlp": 0.01041738,
      "balance_loss_clip": 1.06944108,
      "balance_loss_mlp": 1.02823758,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.6606616051694423,
      "language_loss": 0.87413418,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.8971107,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.6346709728240967
    },
    {
      "auxiliary_loss_clip": 0.01228677,
      "auxiliary_loss_mlp": 0.01054041,
      "balance_loss_clip": 1.06631434,
      "balance_loss_mlp": 1.04061818,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.506621960415769,
      "language_loss": 0.80501938,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.82784653,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.7829411029815674
    },
    {
      "auxiliary_loss_clip": 0.01229908,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.06391716,
      "balance_loss_mlp": 1.02133036,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.7470764607153975,
      "language_loss": 0.9099651,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93261147,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.7301666736602783
    },
    {
      "auxiliary_loss_clip": 0.0124636,
      "auxiliary_loss_mlp": 0.0104853,
      "balance_loss_clip": 1.07178259,
      "balance_loss_mlp": 1.03489256,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.8430344618055092,
      "language_loss": 0.79409564,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81704462,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.6808278560638428
    },
    {
      "auxiliary_loss_clip": 0.01243982,
      "auxiliary_loss_mlp": 0.00905892,
      "balance_loss_clip": 1.06679165,
      "balance_loss_mlp": 1.00048578,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 7.711630833287369,
      "language_loss": 0.81776154,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.83926034,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.729349374771118
    },
    {
      "auxiliary_loss_clip": 0.01268325,
      "auxiliary_loss_mlp": 0.01047958,
      "balance_loss_clip": 1.07347894,
      "balance_loss_mlp": 1.03401017,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 12.082300526980386,
      "language_loss": 0.81888831,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.84205115,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.572754144668579
    },
    {
      "auxiliary_loss_clip": 0.01258005,
      "auxiliary_loss_mlp": 0.00905794,
      "balance_loss_clip": 1.07014394,
      "balance_loss_mlp": 1.00042057,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.8375475299851822,
      "language_loss": 0.74583501,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76747298,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.701674461364746
    },
    {
      "auxiliary_loss_clip": 0.01250527,
      "auxiliary_loss_mlp": 0.01050146,
      "balance_loss_clip": 1.07062817,
      "balance_loss_mlp": 1.03606761,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.255991098195543,
      "language_loss": 0.94355869,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96656543,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 4.63042950630188
    },
    {
      "auxiliary_loss_clip": 0.01252267,
      "auxiliary_loss_mlp": 0.01049105,
      "balance_loss_clip": 1.06857133,
      "balance_loss_mlp": 1.03555048,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.22058406152166,
      "language_loss": 0.68847066,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71148443,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 3.8697633743286133
    },
    {
      "auxiliary_loss_clip": 0.01240895,
      "auxiliary_loss_mlp": 0.01044166,
      "balance_loss_clip": 1.06817853,
      "balance_loss_mlp": 1.02994442,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 6.281211993599223,
      "language_loss": 0.76113403,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78398466,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 3.643493413925171
    },
    {
      "auxiliary_loss_clip": 0.01258765,
      "auxiliary_loss_mlp": 0.01043133,
      "balance_loss_clip": 1.0673821,
      "balance_loss_mlp": 1.02919757,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 1.9961819792549267,
      "language_loss": 0.9200058,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94302475,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.686042547225952
    },
    {
      "auxiliary_loss_clip": 0.01158785,
      "auxiliary_loss_mlp": 0.01010105,
      "balance_loss_clip": 1.0490942,
      "balance_loss_mlp": 1.00538445,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8390804492380753,
      "language_loss": 0.63023299,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65192193,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.3248989582061768
    },
    {
      "auxiliary_loss_clip": 0.01241429,
      "auxiliary_loss_mlp": 0.01044025,
      "balance_loss_clip": 1.06847763,
      "balance_loss_mlp": 1.03026235,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 5.333507204267257,
      "language_loss": 0.85933435,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88218892,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.7233970165252686
    },
    {
      "auxiliary_loss_clip": 0.01273023,
      "auxiliary_loss_mlp": 0.01037459,
      "balance_loss_clip": 1.07271302,
      "balance_loss_mlp": 1.02307022,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 1.8110428388121103,
      "language_loss": 0.74422723,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.76733208,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.6349360942840576
    },
    {
      "auxiliary_loss_clip": 0.01271135,
      "auxiliary_loss_mlp": 0.01054335,
      "balance_loss_clip": 1.07317281,
      "balance_loss_mlp": 1.04011381,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 2.8656104425873865,
      "language_loss": 0.70775807,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73101282,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.624075174331665
    },
    {
      "auxiliary_loss_clip": 0.01222754,
      "auxiliary_loss_mlp": 0.0103952,
      "balance_loss_clip": 1.06230974,
      "balance_loss_mlp": 1.02469087,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 1.9734495514460193,
      "language_loss": 0.7763831,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.79900587,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.729186534881592
    },
    {
      "auxiliary_loss_clip": 0.01226206,
      "auxiliary_loss_mlp": 0.01043577,
      "balance_loss_clip": 1.06802344,
      "balance_loss_mlp": 1.02964783,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 3.248356295759161,
      "language_loss": 0.8049944,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82769227,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 2.771411418914795
    },
    {
      "auxiliary_loss_clip": 0.01240934,
      "auxiliary_loss_mlp": 0.01049101,
      "balance_loss_clip": 1.06335628,
      "balance_loss_mlp": 1.03564191,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 1.9900169318679575,
      "language_loss": 0.77640295,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.79930335,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.6612517833709717
    },
    {
      "auxiliary_loss_clip": 0.01165967,
      "auxiliary_loss_mlp": 0.01006132,
      "balance_loss_clip": 1.04580581,
      "balance_loss_mlp": 1.00145853,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8851830733010809,
      "language_loss": 0.63830286,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66002387,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.1073319911956787
    },
    {
      "auxiliary_loss_clip": 0.01271386,
      "auxiliary_loss_mlp": 0.0104951,
      "balance_loss_clip": 1.0727458,
      "balance_loss_mlp": 1.03541982,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.112396415412947,
      "language_loss": 0.80749524,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.83070421,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 2.7745423316955566
    },
    {
      "auxiliary_loss_clip": 0.01256007,
      "auxiliary_loss_mlp": 0.00904747,
      "balance_loss_clip": 1.07076943,
      "balance_loss_mlp": 1.00041139,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 1.8753614264119207,
      "language_loss": 0.83065134,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85225892,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 2.6031203269958496
    },
    {
      "auxiliary_loss_clip": 0.01259924,
      "auxiliary_loss_mlp": 0.01040373,
      "balance_loss_clip": 1.06775439,
      "balance_loss_mlp": 1.02669954,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 3.0790786262138905,
      "language_loss": 0.87944496,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90244794,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 2.682018518447876
    },
    {
      "auxiliary_loss_clip": 0.01255122,
      "auxiliary_loss_mlp": 0.01043716,
      "balance_loss_clip": 1.06855154,
      "balance_loss_mlp": 1.029006,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.274590647339974,
      "language_loss": 0.78476131,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80774969,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.5922162532806396
    },
    {
      "auxiliary_loss_clip": 0.01245694,
      "auxiliary_loss_mlp": 0.00905191,
      "balance_loss_clip": 1.06906247,
      "balance_loss_mlp": 1.00036907,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 2.274054884000558,
      "language_loss": 0.80581141,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.82732022,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.748640298843384
    },
    {
      "auxiliary_loss_clip": 0.01258339,
      "auxiliary_loss_mlp": 0.01039192,
      "balance_loss_clip": 1.07581067,
      "balance_loss_mlp": 1.02567375,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 2.6189644025332797,
      "language_loss": 0.82538617,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.84836149,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.6329405307769775
    },
    {
      "auxiliary_loss_clip": 0.01210369,
      "auxiliary_loss_mlp": 0.01043057,
      "balance_loss_clip": 1.0593214,
      "balance_loss_mlp": 1.02888262,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 3.5498698309942482,
      "language_loss": 0.66934919,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.69188344,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.747800827026367
    },
    {
      "auxiliary_loss_clip": 0.0126979,
      "auxiliary_loss_mlp": 0.01045409,
      "balance_loss_clip": 1.07202756,
      "balance_loss_mlp": 1.03117597,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 4.413180550476123,
      "language_loss": 0.85064912,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87380111,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.610607624053955
    },
    {
      "auxiliary_loss_clip": 0.01273436,
      "auxiliary_loss_mlp": 0.01050781,
      "balance_loss_clip": 1.07681155,
      "balance_loss_mlp": 1.03774595,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 2.323945077595743,
      "language_loss": 0.84956515,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.87280738,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.6089084148406982
    },
    {
      "auxiliary_loss_clip": 0.01203409,
      "auxiliary_loss_mlp": 0.01040758,
      "balance_loss_clip": 1.0602715,
      "balance_loss_mlp": 1.0269599,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.6673707231482606,
      "language_loss": 0.89343631,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91587794,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.7700107097625732
    },
    {
      "auxiliary_loss_clip": 0.01155576,
      "auxiliary_loss_mlp": 0.01005438,
      "balance_loss_clip": 1.03686452,
      "balance_loss_mlp": 1.00057435,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8138404479731016,
      "language_loss": 0.56998348,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59159362,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.126833200454712
    },
    {
      "auxiliary_loss_clip": 0.01257093,
      "auxiliary_loss_mlp": 0.01045426,
      "balance_loss_clip": 1.06836259,
      "balance_loss_mlp": 1.03271222,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 3.7641040118761437,
      "language_loss": 0.88243127,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90545648,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.783670425415039
    },
    {
      "auxiliary_loss_clip": 0.01256145,
      "auxiliary_loss_mlp": 0.01044779,
      "balance_loss_clip": 1.07429171,
      "balance_loss_mlp": 1.0324285,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 2.7403612752838122,
      "language_loss": 0.78246963,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80547881,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.637096643447876
    },
    {
      "auxiliary_loss_clip": 0.01252684,
      "auxiliary_loss_mlp": 0.01044732,
      "balance_loss_clip": 1.07196927,
      "balance_loss_mlp": 1.03076649,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 3.2596391283553,
      "language_loss": 0.85924512,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.88221931,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.6993653774261475
    },
    {
      "auxiliary_loss_clip": 0.01229904,
      "auxiliary_loss_mlp": 0.01051444,
      "balance_loss_clip": 1.06174886,
      "balance_loss_mlp": 1.0387125,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 1.8634507479360112,
      "language_loss": 0.68291867,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.70573217,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.7235028743743896
    },
    {
      "auxiliary_loss_clip": 0.01259779,
      "auxiliary_loss_mlp": 0.00904765,
      "balance_loss_clip": 1.06729782,
      "balance_loss_mlp": 1.00042808,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 2.0897928492821802,
      "language_loss": 0.73642868,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75807416,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 3.6427958011627197
    },
    {
      "auxiliary_loss_clip": 0.01268488,
      "auxiliary_loss_mlp": 0.01045471,
      "balance_loss_clip": 1.07165098,
      "balance_loss_mlp": 1.03205991,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 2.082143115067748,
      "language_loss": 1.001351,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02449059,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 5.440247297286987
    },
    {
      "auxiliary_loss_clip": 0.01257577,
      "auxiliary_loss_mlp": 0.01045787,
      "balance_loss_clip": 1.07296872,
      "balance_loss_mlp": 1.03240025,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.7855795473873959,
      "language_loss": 0.93559909,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.95863271,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.696333408355713
    },
    {
      "auxiliary_loss_clip": 0.01235604,
      "auxiliary_loss_mlp": 0.01040067,
      "balance_loss_clip": 1.06627512,
      "balance_loss_mlp": 1.02689469,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 2.0026261811303976,
      "language_loss": 0.6774236,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.70018035,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.692570686340332
    },
    {
      "auxiliary_loss_clip": 0.01208595,
      "auxiliary_loss_mlp": 0.01046237,
      "balance_loss_clip": 1.05903983,
      "balance_loss_mlp": 1.03186584,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 1.9231705919423938,
      "language_loss": 0.81214261,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83469093,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.8007702827453613
    },
    {
      "auxiliary_loss_clip": 0.01246547,
      "auxiliary_loss_mlp": 0.01046124,
      "balance_loss_clip": 1.06750894,
      "balance_loss_mlp": 1.03181899,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 2.6369741708958996,
      "language_loss": 0.84022462,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86315137,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.698606491088867
    },
    {
      "auxiliary_loss_clip": 0.01245609,
      "auxiliary_loss_mlp": 0.00905428,
      "balance_loss_clip": 1.07144952,
      "balance_loss_mlp": 1.00046563,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 3.0817052209634075,
      "language_loss": 0.82551861,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84702897,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.717217445373535
    },
    {
      "auxiliary_loss_clip": 0.01144776,
      "auxiliary_loss_mlp": 0.01005054,
      "balance_loss_clip": 1.0349884,
      "balance_loss_mlp": 0.99995214,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7946104139183451,
      "language_loss": 0.62642145,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64791971,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.238133192062378
    },
    {
      "auxiliary_loss_clip": 0.01237876,
      "auxiliary_loss_mlp": 0.01049377,
      "balance_loss_clip": 1.07002831,
      "balance_loss_mlp": 1.03460717,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 3.0068270694911634,
      "language_loss": 0.85578853,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.87866104,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.720633029937744
    },
    {
      "auxiliary_loss_clip": 0.01213502,
      "auxiliary_loss_mlp": 0.01040774,
      "balance_loss_clip": 1.05721664,
      "balance_loss_mlp": 1.02623677,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.1446440237371553,
      "language_loss": 0.81611633,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.83865905,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 2.675173282623291
    },
    {
      "auxiliary_loss_clip": 0.01186614,
      "auxiliary_loss_mlp": 0.01040291,
      "balance_loss_clip": 1.05310965,
      "balance_loss_mlp": 1.02768481,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.9159474966989194,
      "language_loss": 0.84712243,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.8693915,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.800997018814087
    },
    {
      "auxiliary_loss_clip": 0.01263203,
      "auxiliary_loss_mlp": 0.01049619,
      "balance_loss_clip": 1.07319331,
      "balance_loss_mlp": 1.03402615,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 4.6565299285356,
      "language_loss": 0.87865627,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.90178454,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 2.700029134750366
    },
    {
      "auxiliary_loss_clip": 0.01248426,
      "auxiliary_loss_mlp": 0.01040408,
      "balance_loss_clip": 1.07093644,
      "balance_loss_mlp": 1.02774215,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.7281739449017617,
      "language_loss": 0.8667711,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.8896594,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 2.6814839839935303
    },
    {
      "auxiliary_loss_clip": 0.01269664,
      "auxiliary_loss_mlp": 0.01039204,
      "balance_loss_clip": 1.07294822,
      "balance_loss_mlp": 1.02521443,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.5149087682525137,
      "language_loss": 0.96152818,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98461682,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 2.6492438316345215
    },
    {
      "auxiliary_loss_clip": 0.01256269,
      "auxiliary_loss_mlp": 0.009047,
      "balance_loss_clip": 1.07361817,
      "balance_loss_mlp": 1.00045443,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 2.5724037073861448,
      "language_loss": 0.88018918,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.90179896,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.5803635120391846
    },
    {
      "auxiliary_loss_clip": 0.01270415,
      "auxiliary_loss_mlp": 0.01042401,
      "balance_loss_clip": 1.07745743,
      "balance_loss_mlp": 1.02950835,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.651940913899425,
      "language_loss": 0.96455747,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98768568,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.679713726043701
    },
    {
      "auxiliary_loss_clip": 0.01274135,
      "auxiliary_loss_mlp": 0.0090566,
      "balance_loss_clip": 1.0770309,
      "balance_loss_mlp": 1.00046015,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.7461125924517695,
      "language_loss": 0.81406265,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83586061,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.5677456855773926
    },
    {
      "auxiliary_loss_clip": 0.01273784,
      "auxiliary_loss_mlp": 0.01052959,
      "balance_loss_clip": 1.07702184,
      "balance_loss_mlp": 1.03905272,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.455618285111642,
      "language_loss": 0.87442684,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89769423,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.695482015609741
    },
    {
      "auxiliary_loss_clip": 0.01243591,
      "auxiliary_loss_mlp": 0.01040726,
      "balance_loss_clip": 1.06936789,
      "balance_loss_mlp": 1.02685571,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.136587038613382,
      "language_loss": 0.77569437,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79853749,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.689682722091675
    },
    {
      "auxiliary_loss_clip": 0.01247531,
      "auxiliary_loss_mlp": 0.01047727,
      "balance_loss_clip": 1.06966996,
      "balance_loss_mlp": 1.03567505,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.5840842359605236,
      "language_loss": 0.70040226,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72335482,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.7803046703338623
    },
    {
      "auxiliary_loss_clip": 0.01231261,
      "auxiliary_loss_mlp": 0.01055992,
      "balance_loss_clip": 1.06633997,
      "balance_loss_mlp": 1.04106688,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 6.8184651369387685,
      "language_loss": 0.85111749,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87399,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.8423590660095215
    },
    {
      "auxiliary_loss_clip": 0.01258281,
      "auxiliary_loss_mlp": 0.01049581,
      "balance_loss_clip": 1.07224774,
      "balance_loss_mlp": 1.03452444,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.7785633145984847,
      "language_loss": 0.81181008,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.8348887,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.673401117324829
    },
    {
      "auxiliary_loss_clip": 0.01236494,
      "auxiliary_loss_mlp": 0.01043783,
      "balance_loss_clip": 1.06675124,
      "balance_loss_mlp": 1.03136182,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.5339583409154915,
      "language_loss": 0.89964265,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92244542,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.8323001861572266
    },
    {
      "auxiliary_loss_clip": 0.01250689,
      "auxiliary_loss_mlp": 0.01049543,
      "balance_loss_clip": 1.06582582,
      "balance_loss_mlp": 1.0359236,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.061827381974012,
      "language_loss": 0.779293,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80229533,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.691666841506958
    },
    {
      "auxiliary_loss_clip": 0.01247788,
      "auxiliary_loss_mlp": 0.01048454,
      "balance_loss_clip": 1.06918991,
      "balance_loss_mlp": 1.03431606,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.6024462743857133,
      "language_loss": 0.8431266,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86608899,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.767423391342163
    },
    {
      "auxiliary_loss_clip": 0.01253312,
      "auxiliary_loss_mlp": 0.009051,
      "balance_loss_clip": 1.07053363,
      "balance_loss_mlp": 1.00042105,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 2.036105201534222,
      "language_loss": 0.85431135,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.87589544,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.6882989406585693
    },
    {
      "auxiliary_loss_clip": 0.01244181,
      "auxiliary_loss_mlp": 0.01045521,
      "balance_loss_clip": 1.06794906,
      "balance_loss_mlp": 1.02957094,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 2.0569214873793595,
      "language_loss": 0.80804145,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83093858,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.6628170013427734
    },
    {
      "auxiliary_loss_clip": 0.01230369,
      "auxiliary_loss_mlp": 0.00904835,
      "balance_loss_clip": 1.06733894,
      "balance_loss_mlp": 1.0004524,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 1.8531419137780147,
      "language_loss": 0.91609782,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.93744993,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 4.726661205291748
    },
    {
      "auxiliary_loss_clip": 0.012494,
      "auxiliary_loss_mlp": 0.01044944,
      "balance_loss_clip": 1.0732379,
      "balance_loss_mlp": 1.03081799,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.700227821256782,
      "language_loss": 0.76057506,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78351849,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 4.616180181503296
    },
    {
      "auxiliary_loss_clip": 0.01263967,
      "auxiliary_loss_mlp": 0.00905711,
      "balance_loss_clip": 1.07108641,
      "balance_loss_mlp": 1.00048852,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.075282632571393,
      "language_loss": 0.86215138,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88384813,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.720172882080078
    },
    {
      "auxiliary_loss_clip": 0.01271281,
      "auxiliary_loss_mlp": 0.01047378,
      "balance_loss_clip": 1.07495356,
      "balance_loss_mlp": 1.03318572,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.6663638587685077,
      "language_loss": 0.84232736,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86551392,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.671485424041748
    },
    {
      "auxiliary_loss_clip": 0.01241204,
      "auxiliary_loss_mlp": 0.01050187,
      "balance_loss_clip": 1.07138205,
      "balance_loss_mlp": 1.03758025,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.489867775250388,
      "language_loss": 0.86636865,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.88928258,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.7117674350738525
    },
    {
      "auxiliary_loss_clip": 0.01254023,
      "auxiliary_loss_mlp": 0.01044877,
      "balance_loss_clip": 1.07278323,
      "balance_loss_mlp": 1.03123331,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.3647690414631977,
      "language_loss": 0.80063754,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82362652,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.7494866847991943
    },
    {
      "auxiliary_loss_clip": 0.01259054,
      "auxiliary_loss_mlp": 0.01057418,
      "balance_loss_clip": 1.07206297,
      "balance_loss_mlp": 1.04408431,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.3238188390328025,
      "language_loss": 0.73091537,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75408012,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.744856834411621
    },
    {
      "auxiliary_loss_clip": 0.01249847,
      "auxiliary_loss_mlp": 0.01042319,
      "balance_loss_clip": 1.07413685,
      "balance_loss_mlp": 1.02770948,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.382019720230749,
      "language_loss": 0.74851376,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77143538,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.7505276203155518
    },
    {
      "auxiliary_loss_clip": 0.01270306,
      "auxiliary_loss_mlp": 0.01048923,
      "balance_loss_clip": 1.07609081,
      "balance_loss_mlp": 1.03461778,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 3.91344429383039,
      "language_loss": 0.84418368,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.86737591,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 2.6113240718841553
    },
    {
      "auxiliary_loss_clip": 0.01254659,
      "auxiliary_loss_mlp": 0.01055293,
      "balance_loss_clip": 1.07200193,
      "balance_loss_mlp": 1.04150033,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.3593989184417286,
      "language_loss": 0.8477118,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87081128,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.664236545562744
    },
    {
      "auxiliary_loss_clip": 0.01242998,
      "auxiliary_loss_mlp": 0.01044316,
      "balance_loss_clip": 1.07034063,
      "balance_loss_mlp": 1.03061879,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 3.763874655068232,
      "language_loss": 0.91482854,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93770176,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 2.6933023929595947
    },
    {
      "auxiliary_loss_clip": 0.01213692,
      "auxiliary_loss_mlp": 0.00904965,
      "balance_loss_clip": 1.06146669,
      "balance_loss_mlp": 1.00045156,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 2.262502437367757,
      "language_loss": 0.83079052,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85197711,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.7049691677093506
    },
    {
      "auxiliary_loss_clip": 0.01229416,
      "auxiliary_loss_mlp": 0.0104669,
      "balance_loss_clip": 1.06847608,
      "balance_loss_mlp": 1.03332639,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 2.7898390065923677,
      "language_loss": 0.79722726,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.81998831,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.659665584564209
    },
    {
      "auxiliary_loss_clip": 0.01268054,
      "auxiliary_loss_mlp": 0.01046554,
      "balance_loss_clip": 1.07433701,
      "balance_loss_mlp": 1.03298759,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 2.0119735602831534,
      "language_loss": 0.7406671,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.76381314,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.8319613933563232
    },
    {
      "auxiliary_loss_clip": 0.0123865,
      "auxiliary_loss_mlp": 0.0105688,
      "balance_loss_clip": 1.06858242,
      "balance_loss_mlp": 1.04243755,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.353040097802888,
      "language_loss": 0.90993363,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93288887,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.739504337310791
    },
    {
      "auxiliary_loss_clip": 0.01244328,
      "auxiliary_loss_mlp": 0.01045117,
      "balance_loss_clip": 1.0696193,
      "balance_loss_mlp": 1.03188503,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 2.788233260699296,
      "language_loss": 0.83291578,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85581017,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.648845911026001
    },
    {
      "auxiliary_loss_clip": 0.01258834,
      "auxiliary_loss_mlp": 0.01056712,
      "balance_loss_clip": 1.07430935,
      "balance_loss_mlp": 1.04250836,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.640793820443797,
      "language_loss": 0.82224673,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84540212,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.6515610218048096
    },
    {
      "auxiliary_loss_clip": 0.01239923,
      "auxiliary_loss_mlp": 0.01052984,
      "balance_loss_clip": 1.0686419,
      "balance_loss_mlp": 1.03971624,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 3.6527006981846895,
      "language_loss": 0.83029974,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.85322875,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.6419742107391357
    },
    {
      "auxiliary_loss_clip": 0.01231776,
      "auxiliary_loss_mlp": 0.00905674,
      "balance_loss_clip": 1.06584179,
      "balance_loss_mlp": 1.00035596,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.993580897618798,
      "language_loss": 0.80895853,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83033299,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.9018638134002686
    },
    {
      "auxiliary_loss_clip": 0.01221981,
      "auxiliary_loss_mlp": 0.01043227,
      "balance_loss_clip": 1.06880713,
      "balance_loss_mlp": 1.0299114,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.6632126306430475,
      "language_loss": 0.87576324,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89841533,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.7319819927215576
    },
    {
      "auxiliary_loss_clip": 0.0125826,
      "auxiliary_loss_mlp": 0.01051914,
      "balance_loss_clip": 1.07137775,
      "balance_loss_mlp": 1.03906918,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 43.224637382654024,
      "language_loss": 0.82264555,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84574729,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.6426782608032227
    },
    {
      "auxiliary_loss_clip": 0.01266831,
      "auxiliary_loss_mlp": 0.01048226,
      "balance_loss_clip": 1.07187486,
      "balance_loss_mlp": 1.03427887,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 1.9705922699910883,
      "language_loss": 0.74294943,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.76610005,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.5869529247283936
    },
    {
      "auxiliary_loss_clip": 0.01264394,
      "auxiliary_loss_mlp": 0.0090538,
      "balance_loss_clip": 1.07108426,
      "balance_loss_mlp": 1.00036979,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 6.079282377140417,
      "language_loss": 0.62054342,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.64224112,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.652658224105835
    },
    {
      "auxiliary_loss_clip": 0.01114166,
      "auxiliary_loss_mlp": 0.01012014,
      "balance_loss_clip": 1.03107691,
      "balance_loss_mlp": 1.0065304,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0320489353199147,
      "language_loss": 0.59430456,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61556637,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.272125005722046
    },
    {
      "auxiliary_loss_clip": 0.01266945,
      "auxiliary_loss_mlp": 0.01047558,
      "balance_loss_clip": 1.07502818,
      "balance_loss_mlp": 1.03495193,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.207171376599358,
      "language_loss": 0.84267592,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86582088,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.5829925537109375
    },
    {
      "auxiliary_loss_clip": 0.01251035,
      "auxiliary_loss_mlp": 0.01046516,
      "balance_loss_clip": 1.07135904,
      "balance_loss_mlp": 1.03336692,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.152049246631547,
      "language_loss": 0.85409451,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.87707007,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.6958272457122803
    },
    {
      "auxiliary_loss_clip": 0.01168814,
      "auxiliary_loss_mlp": 0.01039275,
      "balance_loss_clip": 1.04850245,
      "balance_loss_mlp": 1.02594721,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.266499651561012,
      "language_loss": 0.94216591,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96424681,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.1635901927948
    },
    {
      "auxiliary_loss_clip": 0.01153353,
      "auxiliary_loss_mlp": 0.01009409,
      "balance_loss_clip": 1.03554249,
      "balance_loss_mlp": 1.00382984,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8306452748335005,
      "language_loss": 0.55423295,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57586062,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 4.140892744064331
    },
    {
      "auxiliary_loss_clip": 0.01235176,
      "auxiliary_loss_mlp": 0.01048143,
      "balance_loss_clip": 1.06306279,
      "balance_loss_mlp": 1.03593636,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.702475988233019,
      "language_loss": 0.81409967,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83693284,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 4.535857439041138
    },
    {
      "auxiliary_loss_clip": 0.01266511,
      "auxiliary_loss_mlp": 0.01050519,
      "balance_loss_clip": 1.07339954,
      "balance_loss_mlp": 1.03583884,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.5637991998364884,
      "language_loss": 0.81673324,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.83990353,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.6747586727142334
    },
    {
      "auxiliary_loss_clip": 0.01242254,
      "auxiliary_loss_mlp": 0.00906031,
      "balance_loss_clip": 1.06960809,
      "balance_loss_mlp": 1.00035048,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 2.424772661177869,
      "language_loss": 0.77718782,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.79867077,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.6665682792663574
    },
    {
      "auxiliary_loss_clip": 0.0125799,
      "auxiliary_loss_mlp": 0.01054977,
      "balance_loss_clip": 1.07570529,
      "balance_loss_mlp": 1.04197705,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.60649207984394,
      "language_loss": 0.81169993,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83482957,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.6721136569976807
    },
    {
      "auxiliary_loss_clip": 0.01230019,
      "auxiliary_loss_mlp": 0.01047914,
      "balance_loss_clip": 1.0648042,
      "balance_loss_mlp": 1.03389537,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 2.1161353926269313,
      "language_loss": 0.78088939,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80366874,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.7393136024475098
    },
    {
      "auxiliary_loss_clip": 0.01242372,
      "auxiliary_loss_mlp": 0.01051048,
      "balance_loss_clip": 1.07194114,
      "balance_loss_mlp": 1.03826261,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.7130027648799584,
      "language_loss": 0.9197889,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.9427231,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.7041642665863037
    },
    {
      "auxiliary_loss_clip": 0.01222508,
      "auxiliary_loss_mlp": 0.01041478,
      "balance_loss_clip": 1.06589413,
      "balance_loss_mlp": 1.02826965,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.6011195287957707,
      "language_loss": 0.80244154,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82508141,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.9489376544952393
    },
    {
      "auxiliary_loss_clip": 0.01241882,
      "auxiliary_loss_mlp": 0.0105041,
      "balance_loss_clip": 1.06765389,
      "balance_loss_mlp": 1.0367012,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 1.9823826611425754,
      "language_loss": 0.82742298,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.85034591,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.688019275665283
    },
    {
      "auxiliary_loss_clip": 0.0125226,
      "auxiliary_loss_mlp": 0.01050465,
      "balance_loss_clip": 1.07079673,
      "balance_loss_mlp": 1.03560567,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 4.278653186453731,
      "language_loss": 0.86977625,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89280355,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 2.6461830139160156
    },
    {
      "auxiliary_loss_clip": 0.01242648,
      "auxiliary_loss_mlp": 0.01044304,
      "balance_loss_clip": 1.07427979,
      "balance_loss_mlp": 1.03046405,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.0535716467110996,
      "language_loss": 0.84026873,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86313826,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.6920127868652344
    },
    {
      "auxiliary_loss_clip": 0.01244455,
      "auxiliary_loss_mlp": 0.01052881,
      "balance_loss_clip": 1.067168,
      "balance_loss_mlp": 1.03939891,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 4.384663916626635,
      "language_loss": 0.84570014,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.8686735,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 2.664961099624634
    },
    {
      "auxiliary_loss_clip": 0.01231982,
      "auxiliary_loss_mlp": 0.01050684,
      "balance_loss_clip": 1.06990767,
      "balance_loss_mlp": 1.03821492,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.3308020975714245,
      "language_loss": 0.74417406,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76700073,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.7073421478271484
    },
    {
      "auxiliary_loss_clip": 0.01236274,
      "auxiliary_loss_mlp": 0.01043287,
      "balance_loss_clip": 1.07119811,
      "balance_loss_mlp": 1.0292325,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.2171762211407113,
      "language_loss": 0.85462785,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87742352,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.7362613677978516
    },
    {
      "auxiliary_loss_clip": 0.01252507,
      "auxiliary_loss_mlp": 0.01045916,
      "balance_loss_clip": 1.07230806,
      "balance_loss_mlp": 1.03243947,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.2374874818200206,
      "language_loss": 0.90981936,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93280363,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.6122312545776367
    },
    {
      "auxiliary_loss_clip": 0.0123867,
      "auxiliary_loss_mlp": 0.01037407,
      "balance_loss_clip": 1.06953931,
      "balance_loss_mlp": 1.02274442,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 2.701165084236426,
      "language_loss": 0.72297198,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74573272,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.9044010639190674
    },
    {
      "auxiliary_loss_clip": 0.01270045,
      "auxiliary_loss_mlp": 0.01052838,
      "balance_loss_clip": 1.07363617,
      "balance_loss_mlp": 1.03828216,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.5408439999259342,
      "language_loss": 0.85879612,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88202494,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.5930135250091553
    },
    {
      "auxiliary_loss_clip": 0.01267984,
      "auxiliary_loss_mlp": 0.01052871,
      "balance_loss_clip": 1.07234979,
      "balance_loss_mlp": 1.03827405,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 3.302896070981226,
      "language_loss": 0.89195329,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91516185,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.616478443145752
    },
    {
      "auxiliary_loss_clip": 0.01127866,
      "auxiliary_loss_mlp": 0.01012476,
      "balance_loss_clip": 1.02909589,
      "balance_loss_mlp": 1.00711179,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9637398762676161,
      "language_loss": 0.63604999,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65745342,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.285285711288452
    },
    {
      "auxiliary_loss_clip": 0.01253111,
      "auxiliary_loss_mlp": 0.01044687,
      "balance_loss_clip": 1.07074857,
      "balance_loss_mlp": 1.03061438,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 3.2053331745351827,
      "language_loss": 0.93149352,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.95447147,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.665619134902954
    },
    {
      "auxiliary_loss_clip": 0.01217953,
      "auxiliary_loss_mlp": 0.01044948,
      "balance_loss_clip": 1.06298733,
      "balance_loss_mlp": 1.03237104,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.064723941548356,
      "language_loss": 0.84970069,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87232959,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.715827465057373
    },
    {
      "auxiliary_loss_clip": 0.01237327,
      "auxiliary_loss_mlp": 0.01045613,
      "balance_loss_clip": 1.06724977,
      "balance_loss_mlp": 1.03189182,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.18078115128263,
      "language_loss": 0.91691333,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.93974274,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.6270456314086914
    },
    {
      "auxiliary_loss_clip": 0.0122572,
      "auxiliary_loss_mlp": 0.01051485,
      "balance_loss_clip": 1.06399119,
      "balance_loss_mlp": 1.03785896,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.8245308749053788,
      "language_loss": 0.71536928,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.7381413,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.7570414543151855
    },
    {
      "auxiliary_loss_clip": 0.01222018,
      "auxiliary_loss_mlp": 0.01043331,
      "balance_loss_clip": 1.06578696,
      "balance_loss_mlp": 1.02976513,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.7265722795612561,
      "language_loss": 0.91114724,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93380082,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.8040695190429688
    },
    {
      "auxiliary_loss_clip": 0.01198398,
      "auxiliary_loss_mlp": 0.00905032,
      "balance_loss_clip": 1.06024504,
      "balance_loss_mlp": 1.00035393,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 3.7857101931567256,
      "language_loss": 1.00701666,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02805114,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.7801456451416016
    },
    {
      "auxiliary_loss_clip": 0.01245213,
      "auxiliary_loss_mlp": 0.01044346,
      "balance_loss_clip": 1.07275558,
      "balance_loss_mlp": 1.03108358,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.402918036959999,
      "language_loss": 0.74447161,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76736718,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.685931921005249
    },
    {
      "auxiliary_loss_clip": 0.01233994,
      "auxiliary_loss_mlp": 0.01046096,
      "balance_loss_clip": 1.06934857,
      "balance_loss_mlp": 1.03071809,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 3.5686582272304332,
      "language_loss": 0.89331889,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91611981,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 3.6995909214019775
    },
    {
      "auxiliary_loss_clip": 0.01262178,
      "auxiliary_loss_mlp": 0.00904927,
      "balance_loss_clip": 1.07464075,
      "balance_loss_mlp": 1.00032258,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 4.492539741668312,
      "language_loss": 0.9073683,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.9290393,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 4.49189829826355
    },
    {
      "auxiliary_loss_clip": 0.01221484,
      "auxiliary_loss_mlp": 0.01048155,
      "balance_loss_clip": 1.06659055,
      "balance_loss_mlp": 1.03459513,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.8982785681612317,
      "language_loss": 0.86640155,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88909793,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.7983815670013428
    },
    {
      "auxiliary_loss_clip": 0.01244258,
      "auxiliary_loss_mlp": 0.01048883,
      "balance_loss_clip": 1.06973433,
      "balance_loss_mlp": 1.03582358,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.268196712937554,
      "language_loss": 0.8923682,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91529965,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.6979219913482666
    },
    {
      "auxiliary_loss_clip": 0.01231156,
      "auxiliary_loss_mlp": 0.01038056,
      "balance_loss_clip": 1.06677175,
      "balance_loss_mlp": 1.02475822,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 1.9704744956508182,
      "language_loss": 0.78141516,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80410731,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.697728157043457
    },
    {
      "auxiliary_loss_clip": 0.01232155,
      "auxiliary_loss_mlp": 0.01050757,
      "balance_loss_clip": 1.06634796,
      "balance_loss_mlp": 1.03754902,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 1.8912074450560314,
      "language_loss": 0.85745823,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88028735,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 2.7022705078125
    },
    {
      "auxiliary_loss_clip": 0.01233742,
      "auxiliary_loss_mlp": 0.0105486,
      "balance_loss_clip": 1.06849408,
      "balance_loss_mlp": 1.04197907,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 2.3381936252730524,
      "language_loss": 0.85605526,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.8789413,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.7510759830474854
    },
    {
      "auxiliary_loss_clip": 0.01220865,
      "auxiliary_loss_mlp": 0.01041766,
      "balance_loss_clip": 1.06390905,
      "balance_loss_mlp": 1.02871871,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 1.988943103035075,
      "language_loss": 0.98967391,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01230013,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.6572210788726807
    },
    {
      "auxiliary_loss_clip": 0.01231611,
      "auxiliary_loss_mlp": 0.01042712,
      "balance_loss_clip": 1.06888807,
      "balance_loss_mlp": 1.02964711,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.4013505776382993,
      "language_loss": 0.75797391,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.78071713,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.756359815597534
    },
    {
      "auxiliary_loss_clip": 0.0122139,
      "auxiliary_loss_mlp": 0.01040637,
      "balance_loss_clip": 1.06571639,
      "balance_loss_mlp": 1.02794743,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.049742559565456,
      "language_loss": 0.75558746,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77820766,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 2.740384101867676
    },
    {
      "auxiliary_loss_clip": 0.01264649,
      "auxiliary_loss_mlp": 0.01046458,
      "balance_loss_clip": 1.07346511,
      "balance_loss_mlp": 1.03316569,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.646031824043897,
      "language_loss": 0.85382259,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87693369,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.6906137466430664
    },
    {
      "auxiliary_loss_clip": 0.0124607,
      "auxiliary_loss_mlp": 0.01040296,
      "balance_loss_clip": 1.06754792,
      "balance_loss_mlp": 1.02662253,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.509216321588391,
      "language_loss": 0.80627346,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.82913721,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 2.642108678817749
    },
    {
      "auxiliary_loss_clip": 0.01255741,
      "auxiliary_loss_mlp": 0.01042252,
      "balance_loss_clip": 1.07190537,
      "balance_loss_mlp": 1.02937758,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.733455641141029,
      "language_loss": 0.91875041,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.94173032,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.709076404571533
    },
    {
      "auxiliary_loss_clip": 0.01239668,
      "auxiliary_loss_mlp": 0.01046513,
      "balance_loss_clip": 1.07178152,
      "balance_loss_mlp": 1.03149223,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.5808597081695344,
      "language_loss": 0.82126242,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.8441242,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.672215223312378
    },
    {
      "auxiliary_loss_clip": 0.01229275,
      "auxiliary_loss_mlp": 0.0104369,
      "balance_loss_clip": 1.06379557,
      "balance_loss_mlp": 1.03060055,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.8796283799857771,
      "language_loss": 0.79087216,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81360179,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.675570487976074
    },
    {
      "auxiliary_loss_clip": 0.01253983,
      "auxiliary_loss_mlp": 0.01041504,
      "balance_loss_clip": 1.07326174,
      "balance_loss_mlp": 1.02859998,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.6055356355012433,
      "language_loss": 0.83247507,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85542995,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.6140780448913574
    },
    {
      "auxiliary_loss_clip": 0.01256316,
      "auxiliary_loss_mlp": 0.01043117,
      "balance_loss_clip": 1.0696857,
      "balance_loss_mlp": 1.02995026,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.281374824872316,
      "language_loss": 0.78761685,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.81061113,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.6693828105926514
    },
    {
      "auxiliary_loss_clip": 0.01234969,
      "auxiliary_loss_mlp": 0.01036315,
      "balance_loss_clip": 1.06377149,
      "balance_loss_mlp": 1.02407801,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 2.2568196184573828,
      "language_loss": 0.85317266,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87588549,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.7088687419891357
    },
    {
      "auxiliary_loss_clip": 0.01260438,
      "auxiliary_loss_mlp": 0.01039008,
      "balance_loss_clip": 1.07203496,
      "balance_loss_mlp": 1.02667618,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.55760997615975,
      "language_loss": 0.91810006,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94109452,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.60607647895813
    },
    {
      "auxiliary_loss_clip": 0.01231219,
      "auxiliary_loss_mlp": 0.01042851,
      "balance_loss_clip": 1.06860912,
      "balance_loss_mlp": 1.03060198,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 3.6055174450448018,
      "language_loss": 0.79066765,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.81340837,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.7059378623962402
    },
    {
      "auxiliary_loss_clip": 0.01249312,
      "auxiliary_loss_mlp": 0.01046989,
      "balance_loss_clip": 1.07022858,
      "balance_loss_mlp": 1.03429317,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.705988805776523,
      "language_loss": 0.92526764,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94823062,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.6163551807403564
    },
    {
      "auxiliary_loss_clip": 0.01220538,
      "auxiliary_loss_mlp": 0.01046106,
      "balance_loss_clip": 1.06754649,
      "balance_loss_mlp": 1.03224778,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 3.6320636989709856,
      "language_loss": 0.734474,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.7571404,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.748109817504883
    },
    {
      "auxiliary_loss_clip": 0.01250169,
      "auxiliary_loss_mlp": 0.01043011,
      "balance_loss_clip": 1.07063937,
      "balance_loss_mlp": 1.03011906,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.144825589438901,
      "language_loss": 0.91587573,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.93880749,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.589975595474243
    },
    {
      "auxiliary_loss_clip": 0.01223518,
      "auxiliary_loss_mlp": 0.01043695,
      "balance_loss_clip": 1.06599712,
      "balance_loss_mlp": 1.03006899,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 1.9708091087477426,
      "language_loss": 0.87637818,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89905024,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.746852397918701
    },
    {
      "auxiliary_loss_clip": 0.01233185,
      "auxiliary_loss_mlp": 0.01048761,
      "balance_loss_clip": 1.06585217,
      "balance_loss_mlp": 1.03500462,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 5.200572975920896,
      "language_loss": 0.85175967,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87457919,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.737194061279297
    },
    {
      "auxiliary_loss_clip": 0.01241467,
      "auxiliary_loss_mlp": 0.0104943,
      "balance_loss_clip": 1.06474733,
      "balance_loss_mlp": 1.03603661,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 2.0962885800050945,
      "language_loss": 0.68461001,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70751894,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.8235368728637695
    },
    {
      "auxiliary_loss_clip": 0.0124841,
      "auxiliary_loss_mlp": 0.01047499,
      "balance_loss_clip": 1.07210469,
      "balance_loss_mlp": 1.03379035,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 4.852738584123681,
      "language_loss": 0.81052351,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83348262,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 3.693419933319092
    },
    {
      "auxiliary_loss_clip": 0.01209433,
      "auxiliary_loss_mlp": 0.01045093,
      "balance_loss_clip": 1.05890059,
      "balance_loss_mlp": 1.03320765,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 4.157517020190837,
      "language_loss": 0.863235,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88578033,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 4.560842514038086
    },
    {
      "auxiliary_loss_clip": 0.01219589,
      "auxiliary_loss_mlp": 0.01052768,
      "balance_loss_clip": 1.064466,
      "balance_loss_mlp": 1.04004192,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.7326507728289553,
      "language_loss": 0.83513689,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85786045,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 3.7382068634033203
    },
    {
      "auxiliary_loss_clip": 0.01244649,
      "auxiliary_loss_mlp": 0.01052813,
      "balance_loss_clip": 1.0733788,
      "balance_loss_mlp": 1.04049897,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 2.768092303138668,
      "language_loss": 0.85832828,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.88130295,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.694809675216675
    },
    {
      "auxiliary_loss_clip": 0.01253624,
      "auxiliary_loss_mlp": 0.0104709,
      "balance_loss_clip": 1.07196736,
      "balance_loss_mlp": 1.03447175,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 7.065199437100352,
      "language_loss": 0.82437003,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84737718,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.644824981689453
    },
    {
      "auxiliary_loss_clip": 0.01235852,
      "auxiliary_loss_mlp": 0.01039918,
      "balance_loss_clip": 1.06731069,
      "balance_loss_mlp": 1.02758539,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.280514465861372,
      "language_loss": 0.82735693,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.85011458,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.632167339324951
    },
    {
      "auxiliary_loss_clip": 0.0123339,
      "auxiliary_loss_mlp": 0.01039241,
      "balance_loss_clip": 1.06613743,
      "balance_loss_mlp": 1.02635467,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.653582786447125,
      "language_loss": 0.86437702,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88710332,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 2.6183359622955322
    },
    {
      "auxiliary_loss_clip": 0.012537,
      "auxiliary_loss_mlp": 0.01049016,
      "balance_loss_clip": 1.0751487,
      "balance_loss_mlp": 1.03624856,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.9138451451918645,
      "language_loss": 0.68468958,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70771676,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 2.6861631870269775
    },
    {
      "auxiliary_loss_clip": 0.0125517,
      "auxiliary_loss_mlp": 0.0104663,
      "balance_loss_clip": 1.07098019,
      "balance_loss_mlp": 1.03396392,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.719969777300829,
      "language_loss": 0.88117349,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90419149,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.6497697830200195
    },
    {
      "auxiliary_loss_clip": 0.01233152,
      "auxiliary_loss_mlp": 0.01040369,
      "balance_loss_clip": 1.06566894,
      "balance_loss_mlp": 1.02846563,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.0294425634905915,
      "language_loss": 0.88360333,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90633851,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 2.658345937728882
    },
    {
      "auxiliary_loss_clip": 0.0126418,
      "auxiliary_loss_mlp": 0.01055079,
      "balance_loss_clip": 1.07457829,
      "balance_loss_mlp": 1.04210293,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.8077111827550603,
      "language_loss": 0.7823891,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80558169,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 2.584156036376953
    },
    {
      "auxiliary_loss_clip": 0.01219418,
      "auxiliary_loss_mlp": 0.01037939,
      "balance_loss_clip": 1.06516027,
      "balance_loss_mlp": 1.02613139,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 2.0135588412516165,
      "language_loss": 0.86767876,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89025241,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.791881799697876
    },
    {
      "auxiliary_loss_clip": 0.01250595,
      "auxiliary_loss_mlp": 0.01040109,
      "balance_loss_clip": 1.07125115,
      "balance_loss_mlp": 1.02756166,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 3.2088630713282655,
      "language_loss": 0.84473878,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86764586,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.646523952484131
    },
    {
      "auxiliary_loss_clip": 0.01227976,
      "auxiliary_loss_mlp": 0.01038524,
      "balance_loss_clip": 1.06961095,
      "balance_loss_mlp": 1.02616787,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 3.016777898010311,
      "language_loss": 0.76506019,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78772515,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.777907609939575
    },
    {
      "auxiliary_loss_clip": 0.0124724,
      "auxiliary_loss_mlp": 0.01039635,
      "balance_loss_clip": 1.07038307,
      "balance_loss_mlp": 1.02667058,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.3256095586101697,
      "language_loss": 0.89401722,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91688609,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.6456403732299805
    },
    {
      "auxiliary_loss_clip": 0.01219322,
      "auxiliary_loss_mlp": 0.01042924,
      "balance_loss_clip": 1.06975281,
      "balance_loss_mlp": 1.03096724,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.0020836679053113,
      "language_loss": 0.83023661,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85285902,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.748661756515503
    },
    {
      "auxiliary_loss_clip": 0.01215646,
      "auxiliary_loss_mlp": 0.01044341,
      "balance_loss_clip": 1.06633496,
      "balance_loss_mlp": 1.03185964,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 3.484680186745517,
      "language_loss": 0.84511518,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.86771506,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.767213821411133
    },
    {
      "auxiliary_loss_clip": 0.01235555,
      "auxiliary_loss_mlp": 0.01042694,
      "balance_loss_clip": 1.06580389,
      "balance_loss_mlp": 1.0308032,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 1.963771903437594,
      "language_loss": 0.93367672,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95645928,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.699716806411743
    },
    {
      "auxiliary_loss_clip": 0.012166,
      "auxiliary_loss_mlp": 0.00905303,
      "balance_loss_clip": 1.06085193,
      "balance_loss_mlp": 1.00060236,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.260051940938014,
      "language_loss": 0.7537393,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77495831,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.812051296234131
    },
    {
      "auxiliary_loss_clip": 0.01209503,
      "auxiliary_loss_mlp": 0.01045653,
      "balance_loss_clip": 1.05972397,
      "balance_loss_mlp": 1.03299928,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.2538997048009017,
      "language_loss": 0.77914834,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80169994,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.6969659328460693
    },
    {
      "auxiliary_loss_clip": 0.012388,
      "auxiliary_loss_mlp": 0.00904902,
      "balance_loss_clip": 1.07094717,
      "balance_loss_mlp": 1.00057006,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.004826013172043,
      "language_loss": 0.88976431,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91120136,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.7069737911224365
    },
    {
      "auxiliary_loss_clip": 0.01235314,
      "auxiliary_loss_mlp": 0.0104844,
      "balance_loss_clip": 1.07085216,
      "balance_loss_mlp": 1.03569651,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.97670220322399,
      "language_loss": 0.63965219,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66248971,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.828997850418091
    },
    {
      "auxiliary_loss_clip": 0.01247174,
      "auxiliary_loss_mlp": 0.01037631,
      "balance_loss_clip": 1.07202423,
      "balance_loss_mlp": 1.02488184,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.153946842610684,
      "language_loss": 0.7858144,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80866253,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6083433628082275
    },
    {
      "auxiliary_loss_clip": 0.01211362,
      "auxiliary_loss_mlp": 0.01043289,
      "balance_loss_clip": 1.06108522,
      "balance_loss_mlp": 1.03161871,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.8857850618617804,
      "language_loss": 0.90718216,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92972863,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.755857229232788
    },
    {
      "auxiliary_loss_clip": 0.01202297,
      "auxiliary_loss_mlp": 0.0103707,
      "balance_loss_clip": 1.06304538,
      "balance_loss_mlp": 1.02500033,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 2.133294248761773,
      "language_loss": 0.86551219,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88790584,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 3.2272181510925293
    },
    {
      "auxiliary_loss_clip": 0.01224676,
      "auxiliary_loss_mlp": 0.0104236,
      "balance_loss_clip": 1.06452656,
      "balance_loss_mlp": 1.02999854,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.304087135782582,
      "language_loss": 0.86351967,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88619006,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.9611551761627197
    },
    {
      "auxiliary_loss_clip": 0.01260864,
      "auxiliary_loss_mlp": 0.01040328,
      "balance_loss_clip": 1.07099581,
      "balance_loss_mlp": 1.02714968,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.396028001728396,
      "language_loss": 0.87790275,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.90091467,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.697986364364624
    },
    {
      "auxiliary_loss_clip": 0.011462,
      "auxiliary_loss_mlp": 0.01019772,
      "balance_loss_clip": 1.04203916,
      "balance_loss_mlp": 1.01576698,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8379721022287558,
      "language_loss": 0.55574441,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57740414,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 4.216982364654541
    },
    {
      "auxiliary_loss_clip": 0.01209374,
      "auxiliary_loss_mlp": 0.01046307,
      "balance_loss_clip": 1.05926621,
      "balance_loss_mlp": 1.0328598,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 2.270464872387728,
      "language_loss": 0.87508881,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89764559,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 5.60940408706665
    },
    {
      "auxiliary_loss_clip": 0.01223099,
      "auxiliary_loss_mlp": 0.0104268,
      "balance_loss_clip": 1.06923521,
      "balance_loss_mlp": 1.02937651,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 2.326564365071022,
      "language_loss": 0.88404536,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90670311,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.7412431240081787
    },
    {
      "auxiliary_loss_clip": 0.01252231,
      "auxiliary_loss_mlp": 0.01045294,
      "balance_loss_clip": 1.07371342,
      "balance_loss_mlp": 1.03268123,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.1423340478612967,
      "language_loss": 0.74168575,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76466107,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.7171967029571533
    },
    {
      "auxiliary_loss_clip": 0.01243777,
      "auxiliary_loss_mlp": 0.01043168,
      "balance_loss_clip": 1.07021272,
      "balance_loss_mlp": 1.030967,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 2.5164127475530087,
      "language_loss": 0.8942529,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91712236,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.601073980331421
    },
    {
      "auxiliary_loss_clip": 0.01137599,
      "auxiliary_loss_mlp": 0.01004364,
      "balance_loss_clip": 1.03609538,
      "balance_loss_mlp": 1.00054932,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8943198999046456,
      "language_loss": 0.6271286,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64854825,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 3.0415022373199463
    },
    {
      "auxiliary_loss_clip": 0.01220307,
      "auxiliary_loss_mlp": 0.01044035,
      "balance_loss_clip": 1.06537056,
      "balance_loss_mlp": 1.03167295,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.37798178149605,
      "language_loss": 0.89131975,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91396314,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.684744358062744
    },
    {
      "auxiliary_loss_clip": 0.01244267,
      "auxiliary_loss_mlp": 0.00904692,
      "balance_loss_clip": 1.06953764,
      "balance_loss_mlp": 1.00063443,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.8400185929277715,
      "language_loss": 0.82274473,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.84423435,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.7771615982055664
    },
    {
      "auxiliary_loss_clip": 0.01124054,
      "auxiliary_loss_mlp": 0.01012309,
      "balance_loss_clip": 1.03515756,
      "balance_loss_mlp": 1.00851774,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8503224940567458,
      "language_loss": 0.61417091,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63553452,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 3.113373279571533
    },
    {
      "auxiliary_loss_clip": 0.01249733,
      "auxiliary_loss_mlp": 0.01044904,
      "balance_loss_clip": 1.06894946,
      "balance_loss_mlp": 1.03148079,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 2.362614898887628,
      "language_loss": 0.82972324,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.8526696,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.7714757919311523
    },
    {
      "auxiliary_loss_clip": 0.01249652,
      "auxiliary_loss_mlp": 0.01044506,
      "balance_loss_clip": 1.06878042,
      "balance_loss_mlp": 1.03192353,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 2.0103685806814426,
      "language_loss": 0.8858701,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.90881169,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.649660110473633
    },
    {
      "auxiliary_loss_clip": 0.01238499,
      "auxiliary_loss_mlp": 0.01050299,
      "balance_loss_clip": 1.07236433,
      "balance_loss_mlp": 1.03735328,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 2.2547292549535194,
      "language_loss": 0.83157432,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85446227,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 2.738919973373413
    },
    {
      "auxiliary_loss_clip": 0.012449,
      "auxiliary_loss_mlp": 0.01038838,
      "balance_loss_clip": 1.0660603,
      "balance_loss_mlp": 1.02658939,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 4.365626391098627,
      "language_loss": 0.86927116,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.8921085,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.6576290130615234
    },
    {
      "auxiliary_loss_clip": 0.01244397,
      "auxiliary_loss_mlp": 0.01048606,
      "balance_loss_clip": 1.06596839,
      "balance_loss_mlp": 1.03554678,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.0599240766574414,
      "language_loss": 0.80994427,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83287436,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.695427179336548
    },
    {
      "auxiliary_loss_clip": 0.0126375,
      "auxiliary_loss_mlp": 0.01046146,
      "balance_loss_clip": 1.07372773,
      "balance_loss_mlp": 1.03161418,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.433135350554184,
      "language_loss": 0.74673384,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76983285,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.591946601867676
    },
    {
      "auxiliary_loss_clip": 0.01231514,
      "auxiliary_loss_mlp": 0.01039118,
      "balance_loss_clip": 1.06350434,
      "balance_loss_mlp": 1.02465785,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 2.134747932406358,
      "language_loss": 0.75915337,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.78185976,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.750762939453125
    },
    {
      "auxiliary_loss_clip": 0.01241817,
      "auxiliary_loss_mlp": 0.01039615,
      "balance_loss_clip": 1.07057571,
      "balance_loss_mlp": 1.02705073,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.7079694745414074,
      "language_loss": 0.87369299,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89650726,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.7218387126922607
    },
    {
      "auxiliary_loss_clip": 0.01229005,
      "auxiliary_loss_mlp": 0.01044936,
      "balance_loss_clip": 1.06854331,
      "balance_loss_mlp": 1.03200173,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.0509421632184295,
      "language_loss": 0.76261294,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78535235,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.665173053741455
    },
    {
      "auxiliary_loss_clip": 0.01130714,
      "auxiliary_loss_mlp": 0.01003044,
      "balance_loss_clip": 1.02758515,
      "balance_loss_mlp": 0.99937195,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.9862919514039986,
      "language_loss": 0.64623779,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66757536,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.1765010356903076
    },
    {
      "auxiliary_loss_clip": 0.01263177,
      "auxiliary_loss_mlp": 0.01043214,
      "balance_loss_clip": 1.07689762,
      "balance_loss_mlp": 1.03143597,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 2.4605471550597664,
      "language_loss": 0.84036154,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86342543,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.649858236312866
    },
    {
      "auxiliary_loss_clip": 0.01126806,
      "auxiliary_loss_mlp": 0.01001582,
      "balance_loss_clip": 1.02543879,
      "balance_loss_mlp": 0.99786252,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8795999361681949,
      "language_loss": 0.64564085,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66692472,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.0552377700805664
    },
    {
      "auxiliary_loss_clip": 0.01241911,
      "auxiliary_loss_mlp": 0.0104004,
      "balance_loss_clip": 1.06617236,
      "balance_loss_mlp": 1.02783906,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 2.675984646406551,
      "language_loss": 0.84219706,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86501658,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.6770451068878174
    },
    {
      "auxiliary_loss_clip": 0.01259046,
      "auxiliary_loss_mlp": 0.01038281,
      "balance_loss_clip": 1.07335281,
      "balance_loss_mlp": 1.02582359,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.164208927256232,
      "language_loss": 0.87127221,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89424551,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.656032085418701
    },
    {
      "auxiliary_loss_clip": 0.01226907,
      "auxiliary_loss_mlp": 0.01051097,
      "balance_loss_clip": 1.0679338,
      "balance_loss_mlp": 1.03796601,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.9054068430774727,
      "language_loss": 0.8058452,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.82862526,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.681036949157715
    },
    {
      "auxiliary_loss_clip": 0.01222411,
      "auxiliary_loss_mlp": 0.01036852,
      "balance_loss_clip": 1.06572199,
      "balance_loss_mlp": 1.02313137,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 1.827973595023984,
      "language_loss": 0.82658291,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.84917557,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.779107093811035
    },
    {
      "auxiliary_loss_clip": 0.01112013,
      "auxiliary_loss_mlp": 0.01006458,
      "balance_loss_clip": 1.02832532,
      "balance_loss_mlp": 1.00202322,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8628922983730591,
      "language_loss": 0.59949547,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62068021,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.3073909282684326
    },
    {
      "auxiliary_loss_clip": 0.01232632,
      "auxiliary_loss_mlp": 0.01055171,
      "balance_loss_clip": 1.06431031,
      "balance_loss_mlp": 1.04256463,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.6584578258682994,
      "language_loss": 0.61044967,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63332772,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 3.6247427463531494
    },
    {
      "auxiliary_loss_clip": 0.01250823,
      "auxiliary_loss_mlp": 0.01047779,
      "balance_loss_clip": 1.07242823,
      "balance_loss_mlp": 1.03521442,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.157981041271462,
      "language_loss": 0.80325502,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.82624102,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 3.5685949325561523
    },
    {
      "auxiliary_loss_clip": 0.01239908,
      "auxiliary_loss_mlp": 0.01040397,
      "balance_loss_clip": 1.07047415,
      "balance_loss_mlp": 1.02678895,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.260119368657514,
      "language_loss": 0.67643154,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.6992346,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 4.576152086257935
    },
    {
      "auxiliary_loss_clip": 0.01237551,
      "auxiliary_loss_mlp": 0.01038314,
      "balance_loss_clip": 1.07108808,
      "balance_loss_mlp": 1.026196,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.428683148605552,
      "language_loss": 0.81687951,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83963817,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.789121389389038
    },
    {
      "auxiliary_loss_clip": 0.01251977,
      "auxiliary_loss_mlp": 0.01048918,
      "balance_loss_clip": 1.07413197,
      "balance_loss_mlp": 1.03499389,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.0903296820450907,
      "language_loss": 0.66935194,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69236088,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.737657308578491
    },
    {
      "auxiliary_loss_clip": 0.01214222,
      "auxiliary_loss_mlp": 0.01048573,
      "balance_loss_clip": 1.06476247,
      "balance_loss_mlp": 1.0354476,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.8587908152786243,
      "language_loss": 0.78068775,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80331564,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 2.7267684936523438
    },
    {
      "auxiliary_loss_clip": 0.01235277,
      "auxiliary_loss_mlp": 0.01045337,
      "balance_loss_clip": 1.0683651,
      "balance_loss_mlp": 1.03255165,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.953764650590773,
      "language_loss": 0.90707922,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.92988539,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 2.6379027366638184
    },
    {
      "auxiliary_loss_clip": 0.01265782,
      "auxiliary_loss_mlp": 0.01054957,
      "balance_loss_clip": 1.07669055,
      "balance_loss_mlp": 1.04067576,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 20.958335298611903,
      "language_loss": 0.89328456,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91649199,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.6595630645751953
    },
    {
      "auxiliary_loss_clip": 0.01247484,
      "auxiliary_loss_mlp": 0.01038073,
      "balance_loss_clip": 1.07295537,
      "balance_loss_mlp": 1.02648544,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 1.8372548756223475,
      "language_loss": 0.82920778,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85206336,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 2.6294894218444824
    },
    {
      "auxiliary_loss_clip": 0.01130751,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.02275884,
      "balance_loss_mlp": 1.02858377,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7476030747459789,
      "language_loss": 0.61184567,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63347697,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.0097265243530273
    },
    {
      "auxiliary_loss_clip": 0.01229915,
      "auxiliary_loss_mlp": 0.01053634,
      "balance_loss_clip": 1.06705618,
      "balance_loss_mlp": 1.04060507,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 2.391775087317892,
      "language_loss": 0.84901172,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87184721,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.7041306495666504
    },
    {
      "auxiliary_loss_clip": 0.01243714,
      "auxiliary_loss_mlp": 0.01042441,
      "balance_loss_clip": 1.07041693,
      "balance_loss_mlp": 1.02950037,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.231306212983981,
      "language_loss": 0.67759812,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70045966,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.6302719116210938
    },
    {
      "auxiliary_loss_clip": 0.01225212,
      "auxiliary_loss_mlp": 0.01049048,
      "balance_loss_clip": 1.06790507,
      "balance_loss_mlp": 1.03580403,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 3.3233639461262094,
      "language_loss": 0.82590801,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84865057,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.6637394428253174
    },
    {
      "auxiliary_loss_clip": 0.0124629,
      "auxiliary_loss_mlp": 0.01050941,
      "balance_loss_clip": 1.07226551,
      "balance_loss_mlp": 1.03863251,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.1336912175163554,
      "language_loss": 0.82912207,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85209441,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.6770763397216797
    },
    {
      "auxiliary_loss_clip": 0.01244084,
      "auxiliary_loss_mlp": 0.01053452,
      "balance_loss_clip": 1.07115912,
      "balance_loss_mlp": 1.03990436,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 1.9804339936648563,
      "language_loss": 0.75773799,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.78071332,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.7721455097198486
    },
    {
      "auxiliary_loss_clip": 0.01234154,
      "auxiliary_loss_mlp": 0.01049776,
      "balance_loss_clip": 1.06599927,
      "balance_loss_mlp": 1.03690171,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 2.151326787460636,
      "language_loss": 0.78082407,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.80366337,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.7288644313812256
    },
    {
      "auxiliary_loss_clip": 0.01200587,
      "auxiliary_loss_mlp": 0.01048243,
      "balance_loss_clip": 1.05792248,
      "balance_loss_mlp": 1.03540373,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.924153483074009,
      "language_loss": 0.72376591,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74625421,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.7471439838409424
    },
    {
      "auxiliary_loss_clip": 0.01251906,
      "auxiliary_loss_mlp": 0.01039148,
      "balance_loss_clip": 1.06873584,
      "balance_loss_mlp": 1.02738857,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 1.8758587954578985,
      "language_loss": 0.86095172,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88386226,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.608062982559204
    },
    {
      "auxiliary_loss_clip": 0.01238815,
      "auxiliary_loss_mlp": 0.0105659,
      "balance_loss_clip": 1.06815732,
      "balance_loss_mlp": 1.04476428,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.60640066555589,
      "language_loss": 0.75629342,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77924746,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.6822574138641357
    },
    {
      "auxiliary_loss_clip": 0.01234651,
      "auxiliary_loss_mlp": 0.01043661,
      "balance_loss_clip": 1.06777191,
      "balance_loss_mlp": 1.03029156,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 5.9690044550061305,
      "language_loss": 0.78771007,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81049323,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.7933638095855713
    },
    {
      "auxiliary_loss_clip": 0.01234443,
      "auxiliary_loss_mlp": 0.01042504,
      "balance_loss_clip": 1.06706071,
      "balance_loss_mlp": 1.03057122,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.2699376733480565,
      "language_loss": 0.795946,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.81871545,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.684122323989868
    },
    {
      "auxiliary_loss_clip": 0.01239926,
      "auxiliary_loss_mlp": 0.01044313,
      "balance_loss_clip": 1.06852651,
      "balance_loss_mlp": 1.03134942,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 7.946181392054247,
      "language_loss": 0.76618904,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.78903151,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.636927366256714
    },
    {
      "auxiliary_loss_clip": 0.01246738,
      "auxiliary_loss_mlp": 0.01041619,
      "balance_loss_clip": 1.07110572,
      "balance_loss_mlp": 1.02916193,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.3860093816106653,
      "language_loss": 0.72431719,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.74720079,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.626420736312866
    },
    {
      "auxiliary_loss_clip": 0.01226874,
      "auxiliary_loss_mlp": 0.01040721,
      "balance_loss_clip": 1.06534863,
      "balance_loss_mlp": 1.02714276,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 2.9060573773989136,
      "language_loss": 0.88570762,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90838361,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.6772327423095703
    },
    {
      "auxiliary_loss_clip": 0.01235893,
      "auxiliary_loss_mlp": 0.01046337,
      "balance_loss_clip": 1.07120109,
      "balance_loss_mlp": 1.03462493,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.168638808913948,
      "language_loss": 0.72748542,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.75030768,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.643336772918701
    },
    {
      "auxiliary_loss_clip": 0.01228897,
      "auxiliary_loss_mlp": 0.01039766,
      "balance_loss_clip": 1.06824386,
      "balance_loss_mlp": 1.0272367,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.9352687358964267,
      "language_loss": 0.83591294,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.8585996,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.6776795387268066
    },
    {
      "auxiliary_loss_clip": 0.01223464,
      "auxiliary_loss_mlp": 0.01039396,
      "balance_loss_clip": 1.06571579,
      "balance_loss_mlp": 1.02724266,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 2.5001128868469187,
      "language_loss": 0.77234119,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.7949698,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 3.571758508682251
    },
    {
      "auxiliary_loss_clip": 0.01220088,
      "auxiliary_loss_mlp": 0.01041424,
      "balance_loss_clip": 1.06075954,
      "balance_loss_mlp": 1.02957416,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.7446588519930426,
      "language_loss": 0.83102679,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85364193,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.724449872970581
    },
    {
      "auxiliary_loss_clip": 0.01127273,
      "auxiliary_loss_mlp": 0.01006236,
      "balance_loss_clip": 1.0244267,
      "balance_loss_mlp": 1.00213575,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7610305575985707,
      "language_loss": 0.58808047,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60941553,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 5.175820589065552
    },
    {
      "auxiliary_loss_clip": 0.01178624,
      "auxiliary_loss_mlp": 0.01047502,
      "balance_loss_clip": 1.05314493,
      "balance_loss_mlp": 1.03614187,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 2.009699148449352,
      "language_loss": 0.88081384,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.9030751,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 4.000766038894653
    },
    {
      "auxiliary_loss_clip": 0.01219427,
      "auxiliary_loss_mlp": 0.01051646,
      "balance_loss_clip": 1.06491899,
      "balance_loss_mlp": 1.03897953,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 2.9445288095724425,
      "language_loss": 0.78709292,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80980361,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.886841297149658
    },
    {
      "auxiliary_loss_clip": 0.01218251,
      "auxiliary_loss_mlp": 0.01045415,
      "balance_loss_clip": 1.06538606,
      "balance_loss_mlp": 1.03320241,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 2.1785025444751684,
      "language_loss": 0.83622336,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85886002,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.6539738178253174
    },
    {
      "auxiliary_loss_clip": 0.01248767,
      "auxiliary_loss_mlp": 0.01042763,
      "balance_loss_clip": 1.07458496,
      "balance_loss_mlp": 1.02904785,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.5185816849105174,
      "language_loss": 0.93933976,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.962255,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 2.5697813034057617
    },
    {
      "auxiliary_loss_clip": 0.0125843,
      "auxiliary_loss_mlp": 0.01036413,
      "balance_loss_clip": 1.07472587,
      "balance_loss_mlp": 1.02469468,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 5.341824342423254,
      "language_loss": 0.7875942,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81054258,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 2.7028167247772217
    },
    {
      "auxiliary_loss_clip": 0.012522,
      "auxiliary_loss_mlp": 0.01052039,
      "balance_loss_clip": 1.07240212,
      "balance_loss_mlp": 1.03944421,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 3.004056749046536,
      "language_loss": 0.8308264,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85386872,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 2.717977523803711
    },
    {
      "auxiliary_loss_clip": 0.01234082,
      "auxiliary_loss_mlp": 0.0103869,
      "balance_loss_clip": 1.06364226,
      "balance_loss_mlp": 1.02621436,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 3.171198769053407,
      "language_loss": 0.72730374,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.75003147,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 2.6290247440338135
    },
    {
      "auxiliary_loss_clip": 0.01256415,
      "auxiliary_loss_mlp": 0.01042288,
      "balance_loss_clip": 1.07291687,
      "balance_loss_mlp": 1.03012824,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.236023470206579,
      "language_loss": 0.80423415,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82722121,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.6141257286071777
    },
    {
      "auxiliary_loss_clip": 0.01234216,
      "auxiliary_loss_mlp": 0.01043101,
      "balance_loss_clip": 1.06511354,
      "balance_loss_mlp": 1.03110886,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 1.7998952166977602,
      "language_loss": 0.89653432,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91930747,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 2.707439422607422
    },
    {
      "auxiliary_loss_clip": 0.01241177,
      "auxiliary_loss_mlp": 0.01046974,
      "balance_loss_clip": 1.06534481,
      "balance_loss_mlp": 1.03428447,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 1.8361060568401926,
      "language_loss": 0.84593457,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86881614,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.657217502593994
    },
    {
      "auxiliary_loss_clip": 0.01236364,
      "auxiliary_loss_mlp": 0.01041762,
      "balance_loss_clip": 1.06887293,
      "balance_loss_mlp": 1.02975798,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.6629707067804902,
      "language_loss": 0.8955642,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.91834545,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.710905075073242
    },
    {
      "auxiliary_loss_clip": 0.01242172,
      "auxiliary_loss_mlp": 0.00904608,
      "balance_loss_clip": 1.07098997,
      "balance_loss_mlp": 1.00068855,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.2398473956706515,
      "language_loss": 0.82290149,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84436929,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.6149189472198486
    },
    {
      "auxiliary_loss_clip": 0.01237844,
      "auxiliary_loss_mlp": 0.01046889,
      "balance_loss_clip": 1.06787765,
      "balance_loss_mlp": 1.03283477,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 2.1229727421338365,
      "language_loss": 0.77334285,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.7961902,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.6394262313842773
    },
    {
      "auxiliary_loss_clip": 0.0123844,
      "auxiliary_loss_mlp": 0.00904525,
      "balance_loss_clip": 1.07203794,
      "balance_loss_mlp": 1.00053358,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.64136838834115,
      "language_loss": 0.74570525,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.7671349,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.6729178428649902
    },
    {
      "auxiliary_loss_clip": 0.01259981,
      "auxiliary_loss_mlp": 0.01046904,
      "balance_loss_clip": 1.07465911,
      "balance_loss_mlp": 1.03352928,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 2.271092029676713,
      "language_loss": 0.89363718,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91670597,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.6443586349487305
    },
    {
      "auxiliary_loss_clip": 0.01242043,
      "auxiliary_loss_mlp": 0.01049404,
      "balance_loss_clip": 1.07081556,
      "balance_loss_mlp": 1.03775191,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.8418900074980784,
      "language_loss": 0.78675646,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80967093,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.692042827606201
    },
    {
      "auxiliary_loss_clip": 0.01224786,
      "auxiliary_loss_mlp": 0.01048478,
      "balance_loss_clip": 1.0661335,
      "balance_loss_mlp": 1.03551435,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.846486772050552,
      "language_loss": 0.71597874,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.7387113,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.701857328414917
    },
    {
      "auxiliary_loss_clip": 0.01235851,
      "auxiliary_loss_mlp": 0.00905456,
      "balance_loss_clip": 1.06568933,
      "balance_loss_mlp": 1.00061929,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 2.3966112606906944,
      "language_loss": 0.83982044,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.86123347,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.6231021881103516
    },
    {
      "auxiliary_loss_clip": 0.01203376,
      "auxiliary_loss_mlp": 0.01039605,
      "balance_loss_clip": 1.05790448,
      "balance_loss_mlp": 1.0267483,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 5.2052541166140465,
      "language_loss": 0.74714816,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.76957804,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.740288734436035
    },
    {
      "auxiliary_loss_clip": 0.01243547,
      "auxiliary_loss_mlp": 0.01044198,
      "balance_loss_clip": 1.06977475,
      "balance_loss_mlp": 1.0322175,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.0883846253896805,
      "language_loss": 0.81998563,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.84286302,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.608018159866333
    },
    {
      "auxiliary_loss_clip": 0.01254549,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.07104945,
      "balance_loss_mlp": 1.02155554,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 7.852049386073862,
      "language_loss": 0.90927541,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93216014,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.546295642852783
    },
    {
      "auxiliary_loss_clip": 0.01233277,
      "auxiliary_loss_mlp": 0.01042023,
      "balance_loss_clip": 1.06924343,
      "balance_loss_mlp": 1.0301373,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 3.7854997244621273,
      "language_loss": 0.8566854,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87943834,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.9037418365478516
    },
    {
      "auxiliary_loss_clip": 0.012128,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.06863153,
      "balance_loss_mlp": 1.03354406,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.4273171221517966,
      "language_loss": 0.75710243,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.77968597,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.7516846656799316
    },
    {
      "auxiliary_loss_clip": 0.01227598,
      "auxiliary_loss_mlp": 0.01042729,
      "balance_loss_clip": 1.06496441,
      "balance_loss_mlp": 1.03042078,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 3.424674762391775,
      "language_loss": 0.94793344,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.97063673,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.661456823348999
    },
    {
      "auxiliary_loss_clip": 0.01207509,
      "auxiliary_loss_mlp": 0.01043653,
      "balance_loss_clip": 1.0580976,
      "balance_loss_mlp": 1.02958632,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.8803959403059536,
      "language_loss": 0.88517457,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90768623,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 3.6899163722991943
    },
    {
      "auxiliary_loss_clip": 0.01206345,
      "auxiliary_loss_mlp": 0.0104908,
      "balance_loss_clip": 1.05950356,
      "balance_loss_mlp": 1.0373497,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.5096838547107816,
      "language_loss": 0.80288994,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82544422,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 4.581717014312744
    },
    {
      "auxiliary_loss_clip": 0.01243677,
      "auxiliary_loss_mlp": 0.01046109,
      "balance_loss_clip": 1.06926894,
      "balance_loss_mlp": 1.03302026,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 11.861459605500807,
      "language_loss": 0.70470691,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72760475,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 3.642914056777954
    },
    {
      "auxiliary_loss_clip": 0.01223493,
      "auxiliary_loss_mlp": 0.01035292,
      "balance_loss_clip": 1.0632453,
      "balance_loss_mlp": 1.02339447,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 2.127874540905721,
      "language_loss": 0.85138738,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87397522,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.7435457706451416
    },
    {
      "auxiliary_loss_clip": 0.01106931,
      "auxiliary_loss_mlp": 0.01003256,
      "balance_loss_clip": 1.02313232,
      "balance_loss_mlp": 0.99989414,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8779313002029034,
      "language_loss": 0.61613393,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63723576,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.19384765625
    },
    {
      "auxiliary_loss_clip": 0.0123886,
      "auxiliary_loss_mlp": 0.01041002,
      "balance_loss_clip": 1.06821132,
      "balance_loss_mlp": 1.0282464,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 1.9102587821288675,
      "language_loss": 0.92176139,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94455993,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 2.749403953552246
    },
    {
      "auxiliary_loss_clip": 0.01251638,
      "auxiliary_loss_mlp": 0.01044596,
      "balance_loss_clip": 1.07402217,
      "balance_loss_mlp": 1.03174484,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 2.742075962040118,
      "language_loss": 0.85729551,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88025779,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 2.6310582160949707
    },
    {
      "auxiliary_loss_clip": 0.01218094,
      "auxiliary_loss_mlp": 0.0103321,
      "balance_loss_clip": 1.06415296,
      "balance_loss_mlp": 1.0211159,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 2.0192164758073314,
      "language_loss": 0.80524814,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82776123,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 2.7150790691375732
    },
    {
      "auxiliary_loss_clip": 0.01253114,
      "auxiliary_loss_mlp": 0.01044665,
      "balance_loss_clip": 1.07621312,
      "balance_loss_mlp": 1.03176093,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 2.8666011298152645,
      "language_loss": 0.83970499,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86268282,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 2.6294844150543213
    },
    {
      "auxiliary_loss_clip": 0.01259814,
      "auxiliary_loss_mlp": 0.01048835,
      "balance_loss_clip": 1.07537317,
      "balance_loss_mlp": 1.03690839,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.1074690803603366,
      "language_loss": 0.84122396,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.86431044,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.5657668113708496
    },
    {
      "auxiliary_loss_clip": 0.01242429,
      "auxiliary_loss_mlp": 0.01043676,
      "balance_loss_clip": 1.07147098,
      "balance_loss_mlp": 1.03188586,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 1.9583947264551378,
      "language_loss": 0.87988925,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90275037,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.694721221923828
    },
    {
      "auxiliary_loss_clip": 0.01213469,
      "auxiliary_loss_mlp": 0.01046295,
      "balance_loss_clip": 1.06495512,
      "balance_loss_mlp": 1.03431463,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.0650909616019066,
      "language_loss": 0.92145884,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94405651,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.685349464416504
    },
    {
      "auxiliary_loss_clip": 0.01129552,
      "auxiliary_loss_mlp": 0.01010024,
      "balance_loss_clip": 1.0230515,
      "balance_loss_mlp": 1.00690079,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8399622404832778,
      "language_loss": 0.61220527,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63360101,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.2866482734680176
    },
    {
      "auxiliary_loss_clip": 0.01230944,
      "auxiliary_loss_mlp": 0.01047433,
      "balance_loss_clip": 1.06657028,
      "balance_loss_mlp": 1.03237653,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.3789159374471023,
      "language_loss": 0.73539484,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.75817865,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.6502676010131836
    },
    {
      "auxiliary_loss_clip": 0.01235874,
      "auxiliary_loss_mlp": 0.01048548,
      "balance_loss_clip": 1.06743705,
      "balance_loss_mlp": 1.03560209,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.8094391124559928,
      "language_loss": 0.8299585,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.85280269,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.6432836055755615
    },
    {
      "auxiliary_loss_clip": 0.01257011,
      "auxiliary_loss_mlp": 0.01049216,
      "balance_loss_clip": 1.07622814,
      "balance_loss_mlp": 1.03676438,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 2.045168738681823,
      "language_loss": 0.87918508,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90224737,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.7204272747039795
    },
    {
      "auxiliary_loss_clip": 0.01229193,
      "auxiliary_loss_mlp": 0.01037285,
      "balance_loss_clip": 1.06263566,
      "balance_loss_mlp": 1.02583504,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.7829174260809577,
      "language_loss": 0.84821582,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.8708806,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.723029375076294
    },
    {
      "auxiliary_loss_clip": 0.01244672,
      "auxiliary_loss_mlp": 0.01049916,
      "balance_loss_clip": 1.06929827,
      "balance_loss_mlp": 1.03757155,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.680724105827491,
      "language_loss": 0.74271512,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.765661,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.6930527687072754
    },
    {
      "auxiliary_loss_clip": 0.01203781,
      "auxiliary_loss_mlp": 0.01046778,
      "balance_loss_clip": 1.05868924,
      "balance_loss_mlp": 1.03440428,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 2.019102072556148,
      "language_loss": 0.73177564,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75428128,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.682809591293335
    },
    {
      "auxiliary_loss_clip": 0.01101571,
      "auxiliary_loss_mlp": 0.01006308,
      "balance_loss_clip": 1.01796079,
      "balance_loss_mlp": 1.00292242,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 5.481881262050938,
      "language_loss": 0.63604403,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65712285,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.3004846572875977
    },
    {
      "auxiliary_loss_clip": 0.01237254,
      "auxiliary_loss_mlp": 0.01045284,
      "balance_loss_clip": 1.0678587,
      "balance_loss_mlp": 1.0325464,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 1.9138387730445432,
      "language_loss": 0.8516928,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87451816,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.6840250492095947
    },
    {
      "auxiliary_loss_clip": 0.012471,
      "auxiliary_loss_mlp": 0.01039271,
      "balance_loss_clip": 1.07268381,
      "balance_loss_mlp": 1.02670074,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 1.852589025974892,
      "language_loss": 0.85087532,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.873739,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.6248204708099365
    },
    {
      "auxiliary_loss_clip": 0.01235141,
      "auxiliary_loss_mlp": 0.01048635,
      "balance_loss_clip": 1.07022667,
      "balance_loss_mlp": 1.03475904,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 2.184809088661162,
      "language_loss": 0.7667371,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78957486,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.6500492095947266
    },
    {
      "auxiliary_loss_clip": 0.01251595,
      "auxiliary_loss_mlp": 0.01046366,
      "balance_loss_clip": 1.06796885,
      "balance_loss_mlp": 1.03359842,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.8398854025490743,
      "language_loss": 0.78566051,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80864012,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.5940487384796143
    },
    {
      "auxiliary_loss_clip": 0.01236147,
      "auxiliary_loss_mlp": 0.01046692,
      "balance_loss_clip": 1.0620718,
      "balance_loss_mlp": 1.03446722,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.7720597494781742,
      "language_loss": 0.79794544,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82077384,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.6696674823760986
    },
    {
      "auxiliary_loss_clip": 0.01234095,
      "auxiliary_loss_mlp": 0.01043559,
      "balance_loss_clip": 1.06821942,
      "balance_loss_mlp": 1.03135192,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 1.9028114580835207,
      "language_loss": 0.84243703,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86521357,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.6507740020751953
    },
    {
      "auxiliary_loss_clip": 0.01240199,
      "auxiliary_loss_mlp": 0.00905814,
      "balance_loss_clip": 1.07213569,
      "balance_loss_mlp": 1.00035644,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 3.144583159999923,
      "language_loss": 0.68339795,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70485806,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.6649537086486816
    },
    {
      "auxiliary_loss_clip": 0.01187158,
      "auxiliary_loss_mlp": 0.00905291,
      "balance_loss_clip": 1.06248248,
      "balance_loss_mlp": 1.00033545,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 2.006655313398708,
      "language_loss": 0.83773315,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.85865766,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 4.764115571975708
    },
    {
      "auxiliary_loss_clip": 0.01241369,
      "auxiliary_loss_mlp": 0.01041481,
      "balance_loss_clip": 1.07018781,
      "balance_loss_mlp": 1.02954209,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 4.344564040924097,
      "language_loss": 0.83574808,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85857654,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 3.6080820560455322
    },
    {
      "auxiliary_loss_clip": 0.01220807,
      "auxiliary_loss_mlp": 0.01041449,
      "balance_loss_clip": 1.06405497,
      "balance_loss_mlp": 1.0289259,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.9005962076711962,
      "language_loss": 0.88383472,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90645725,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.7178683280944824
    },
    {
      "auxiliary_loss_clip": 0.01231758,
      "auxiliary_loss_mlp": 0.01044894,
      "balance_loss_clip": 1.07013655,
      "balance_loss_mlp": 1.03353941,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.8925041501757613,
      "language_loss": 0.81163549,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83440202,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.562727928161621
    },
    {
      "auxiliary_loss_clip": 0.01241548,
      "auxiliary_loss_mlp": 0.01049358,
      "balance_loss_clip": 1.06748259,
      "balance_loss_mlp": 1.03611422,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.9117720998978245,
      "language_loss": 0.74133974,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76424873,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 2.553493022918701
    },
    {
      "auxiliary_loss_clip": 0.01231034,
      "auxiliary_loss_mlp": 0.01043639,
      "balance_loss_clip": 1.06927848,
      "balance_loss_mlp": 1.03164649,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 2.2855223004110297,
      "language_loss": 0.78409582,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80684257,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 2.5177009105682373
    },
    {
      "auxiliary_loss_clip": 0.01238579,
      "auxiliary_loss_mlp": 0.01042895,
      "balance_loss_clip": 1.0633204,
      "balance_loss_mlp": 1.02980614,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 3.531906272451568,
      "language_loss": 0.83088386,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85369861,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.628221035003662
    },
    {
      "auxiliary_loss_clip": 0.0123088,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.06621623,
      "balance_loss_mlp": 1.0228951,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.8610054390249624,
      "language_loss": 0.8139931,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83665395,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 2.6219863891601562
    },
    {
      "auxiliary_loss_clip": 0.01242978,
      "auxiliary_loss_mlp": 0.01041272,
      "balance_loss_clip": 1.06557071,
      "balance_loss_mlp": 1.02824247,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 3.2188110049143153,
      "language_loss": 0.83436251,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85720503,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.594006061553955
    },
    {
      "auxiliary_loss_clip": 0.01239995,
      "auxiliary_loss_mlp": 0.00905271,
      "balance_loss_clip": 1.06768572,
      "balance_loss_mlp": 1.00009537,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 1.9720007997306586,
      "language_loss": 0.90430462,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.92575729,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.61319637298584
    },
    {
      "auxiliary_loss_clip": 0.01245939,
      "auxiliary_loss_mlp": 0.01043667,
      "balance_loss_clip": 1.06918788,
      "balance_loss_mlp": 1.03100681,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.2278671173383304,
      "language_loss": 0.75503761,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77793372,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.5943763256073
    },
    {
      "auxiliary_loss_clip": 0.01250786,
      "auxiliary_loss_mlp": 0.01046684,
      "balance_loss_clip": 1.07069635,
      "balance_loss_mlp": 1.03543627,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.832298765439669,
      "language_loss": 0.78737491,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81034958,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.594944477081299
    },
    {
      "auxiliary_loss_clip": 0.01220402,
      "auxiliary_loss_mlp": 0.00904903,
      "balance_loss_clip": 1.06192827,
      "balance_loss_mlp": 1.00012898,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.7987615725909576,
      "language_loss": 0.86404252,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88529557,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.67815899848938
    },
    {
      "auxiliary_loss_clip": 0.01242662,
      "auxiliary_loss_mlp": 0.01042416,
      "balance_loss_clip": 1.06719923,
      "balance_loss_mlp": 1.03030992,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 1.959343752594375,
      "language_loss": 0.90459061,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92744136,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.5944626331329346
    },
    {
      "auxiliary_loss_clip": 0.01227461,
      "auxiliary_loss_mlp": 0.01041688,
      "balance_loss_clip": 1.06530952,
      "balance_loss_mlp": 1.03011835,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.95491819956191,
      "language_loss": 0.84447205,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.8671636,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.682760238647461
    },
    {
      "auxiliary_loss_clip": 0.01210108,
      "auxiliary_loss_mlp": 0.01042173,
      "balance_loss_clip": 1.0591197,
      "balance_loss_mlp": 1.02925074,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.147106066747858,
      "language_loss": 0.83765417,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.86017704,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.6817188262939453
    },
    {
      "auxiliary_loss_clip": 0.01213099,
      "auxiliary_loss_mlp": 0.01048673,
      "balance_loss_clip": 1.05837035,
      "balance_loss_mlp": 1.03467774,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 3.0855267444456973,
      "language_loss": 0.75694168,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.77955937,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.6709530353546143
    },
    {
      "auxiliary_loss_clip": 0.01220401,
      "auxiliary_loss_mlp": 0.01037262,
      "balance_loss_clip": 1.06150389,
      "balance_loss_mlp": 1.02520418,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 5.056154428909846,
      "language_loss": 0.89654469,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.91912138,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.623246908187866
    },
    {
      "auxiliary_loss_clip": 0.01253557,
      "auxiliary_loss_mlp": 0.01045565,
      "balance_loss_clip": 1.06751156,
      "balance_loss_mlp": 1.03366208,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.4787336168519603,
      "language_loss": 0.76131088,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78430212,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.5469038486480713
    },
    {
      "auxiliary_loss_clip": 0.0124667,
      "auxiliary_loss_mlp": 0.00905719,
      "balance_loss_clip": 1.07584286,
      "balance_loss_mlp": 1.00001979,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.9413477435404016,
      "language_loss": 0.76022017,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78174406,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.645090341567993
    },
    {
      "auxiliary_loss_clip": 0.01241457,
      "auxiliary_loss_mlp": 0.01044339,
      "balance_loss_clip": 1.07136321,
      "balance_loss_mlp": 1.03163099,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 1.9701373775960775,
      "language_loss": 0.71894252,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.74180049,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.725071430206299
    },
    {
      "auxiliary_loss_clip": 0.01228084,
      "auxiliary_loss_mlp": 0.01050306,
      "balance_loss_clip": 1.06438613,
      "balance_loss_mlp": 1.03828979,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.0988722416385195,
      "language_loss": 0.92377049,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.9465543,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.667778968811035
    },
    {
      "auxiliary_loss_clip": 0.01236464,
      "auxiliary_loss_mlp": 0.01047419,
      "balance_loss_clip": 1.06507742,
      "balance_loss_mlp": 1.03459764,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 4.2406515104028,
      "language_loss": 0.69379365,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71663249,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.6201891899108887
    },
    {
      "auxiliary_loss_clip": 0.01240751,
      "auxiliary_loss_mlp": 0.01041513,
      "balance_loss_clip": 1.06929684,
      "balance_loss_mlp": 1.02816105,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.2624238055887034,
      "language_loss": 0.74820668,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77102935,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.6025917530059814
    },
    {
      "auxiliary_loss_clip": 0.01240858,
      "auxiliary_loss_mlp": 0.0104437,
      "balance_loss_clip": 1.06872535,
      "balance_loss_mlp": 1.03128707,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.902201666542693,
      "language_loss": 0.74572849,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76858073,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.6174819469451904
    },
    {
      "auxiliary_loss_clip": 0.01230236,
      "auxiliary_loss_mlp": 0.0104658,
      "balance_loss_clip": 1.06659937,
      "balance_loss_mlp": 1.03372288,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.0189733918647197,
      "language_loss": 0.86311901,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88588715,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.6110901832580566
    },
    {
      "auxiliary_loss_clip": 0.01221915,
      "auxiliary_loss_mlp": 0.01038939,
      "balance_loss_clip": 1.06656158,
      "balance_loss_mlp": 1.02697015,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.1567981936436773,
      "language_loss": 0.79917848,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.821787,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 3.569554567337036
    },
    {
      "auxiliary_loss_clip": 0.0122924,
      "auxiliary_loss_mlp": 0.01047853,
      "balance_loss_clip": 1.06542492,
      "balance_loss_mlp": 1.03549683,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 2.4048995770067885,
      "language_loss": 0.86452365,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88729453,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 5.4337780475616455
    },
    {
      "auxiliary_loss_clip": 0.01233317,
      "auxiliary_loss_mlp": 0.01047994,
      "balance_loss_clip": 1.06940174,
      "balance_loss_mlp": 1.03563237,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.038050032361976,
      "language_loss": 0.83588493,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85869807,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 2.657473564147949
    },
    {
      "auxiliary_loss_clip": 0.01230689,
      "auxiliary_loss_mlp": 0.0090553,
      "balance_loss_clip": 1.06657517,
      "balance_loss_mlp": 0.99996793,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.522600658256954,
      "language_loss": 0.85427272,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.87563497,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.7546937465667725
    },
    {
      "auxiliary_loss_clip": 0.01201202,
      "auxiliary_loss_mlp": 0.01048724,
      "balance_loss_clip": 1.06242585,
      "balance_loss_mlp": 1.03566432,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 1.9183638631877789,
      "language_loss": 0.85202897,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87452829,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.691953659057617
    },
    {
      "auxiliary_loss_clip": 0.01236776,
      "auxiliary_loss_mlp": 0.01046303,
      "balance_loss_clip": 1.06710148,
      "balance_loss_mlp": 1.03469801,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 2.042839014013923,
      "language_loss": 0.79127812,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81410885,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.640317916870117
    },
    {
      "auxiliary_loss_clip": 0.01232146,
      "auxiliary_loss_mlp": 0.01042282,
      "balance_loss_clip": 1.06571293,
      "balance_loss_mlp": 1.03004503,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.349925604984209,
      "language_loss": 0.75184035,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77458459,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 2.6261608600616455
    },
    {
      "auxiliary_loss_clip": 0.01210195,
      "auxiliary_loss_mlp": 0.01040704,
      "balance_loss_clip": 1.06284976,
      "balance_loss_mlp": 1.02896202,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 2.3990030736872687,
      "language_loss": 0.80986094,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83236992,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 2.804314613342285
    },
    {
      "auxiliary_loss_clip": 0.01226828,
      "auxiliary_loss_mlp": 0.00904073,
      "balance_loss_clip": 1.06724823,
      "balance_loss_mlp": 0.99998009,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.31682779390473,
      "language_loss": 0.84407502,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.8653841,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.628936767578125
    },
    {
      "auxiliary_loss_clip": 0.01233129,
      "auxiliary_loss_mlp": 0.01051189,
      "balance_loss_clip": 1.06637657,
      "balance_loss_mlp": 1.0375874,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 2.0584491635641684,
      "language_loss": 0.80715245,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82999563,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.7558767795562744
    },
    {
      "auxiliary_loss_clip": 0.01225032,
      "auxiliary_loss_mlp": 0.01044018,
      "balance_loss_clip": 1.06486654,
      "balance_loss_mlp": 1.03058326,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.3625365272866174,
      "language_loss": 0.90106845,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92375898,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.637660503387451
    },
    {
      "auxiliary_loss_clip": 0.01259903,
      "auxiliary_loss_mlp": 0.01047958,
      "balance_loss_clip": 1.07537842,
      "balance_loss_mlp": 1.03567958,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.1378363886597245,
      "language_loss": 0.90577364,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.9288522,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.64109468460083
    },
    {
      "auxiliary_loss_clip": 0.01223444,
      "auxiliary_loss_mlp": 0.0104792,
      "balance_loss_clip": 1.06248164,
      "balance_loss_mlp": 1.03549862,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.6243496760776202,
      "language_loss": 0.83671439,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85942805,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.6466593742370605
    },
    {
      "auxiliary_loss_clip": 0.01237385,
      "auxiliary_loss_mlp": 0.01042723,
      "balance_loss_clip": 1.06911993,
      "balance_loss_mlp": 1.03050375,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.9179001912811793,
      "language_loss": 0.67595196,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.698753,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.664926528930664
    },
    {
      "auxiliary_loss_clip": 0.01244581,
      "auxiliary_loss_mlp": 0.01043981,
      "balance_loss_clip": 1.06704521,
      "balance_loss_mlp": 1.03273344,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 2.3815548988760997,
      "language_loss": 0.799016,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82190168,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.7350285053253174
    },
    {
      "auxiliary_loss_clip": 0.01245844,
      "auxiliary_loss_mlp": 0.01041563,
      "balance_loss_clip": 1.07183516,
      "balance_loss_mlp": 1.02964187,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.6824295589350355,
      "language_loss": 0.82850456,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.85137868,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.587202787399292
    },
    {
      "auxiliary_loss_clip": 0.01243172,
      "auxiliary_loss_mlp": 0.01039477,
      "balance_loss_clip": 1.06713378,
      "balance_loss_mlp": 1.02747822,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 4.322465836593747,
      "language_loss": 0.84683025,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.86965674,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.6580798625946045
    },
    {
      "auxiliary_loss_clip": 0.0124068,
      "auxiliary_loss_mlp": 0.0103804,
      "balance_loss_clip": 1.06943834,
      "balance_loss_mlp": 1.02567816,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 2.014365707329037,
      "language_loss": 0.85120213,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87398928,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.6125307083129883
    },
    {
      "auxiliary_loss_clip": 0.01243563,
      "auxiliary_loss_mlp": 0.01044992,
      "balance_loss_clip": 1.07031119,
      "balance_loss_mlp": 1.03254092,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 1.8456660067201522,
      "language_loss": 0.84889621,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87178171,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.6520698070526123
    },
    {
      "auxiliary_loss_clip": 0.01222801,
      "auxiliary_loss_mlp": 0.01036145,
      "balance_loss_clip": 1.06507468,
      "balance_loss_mlp": 1.02356291,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 2.772551068103177,
      "language_loss": 0.74528682,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.76787627,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.6363601684570312
    },
    {
      "auxiliary_loss_clip": 0.01248659,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.07097816,
      "balance_loss_mlp": 1.02517152,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 3.928795642125696,
      "language_loss": 0.66130573,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68416584,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.600412130355835
    },
    {
      "auxiliary_loss_clip": 0.01200954,
      "auxiliary_loss_mlp": 0.01038985,
      "balance_loss_clip": 1.05511642,
      "balance_loss_mlp": 1.02793384,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.2557457632122273,
      "language_loss": 0.89847457,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92087394,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.701864719390869
    },
    {
      "auxiliary_loss_clip": 0.01257421,
      "auxiliary_loss_mlp": 0.01045851,
      "balance_loss_clip": 1.07445276,
      "balance_loss_mlp": 1.03343523,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.4503050139358074,
      "language_loss": 0.77161956,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79465222,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.5916712284088135
    },
    {
      "auxiliary_loss_clip": 0.01234933,
      "auxiliary_loss_mlp": 0.01043547,
      "balance_loss_clip": 1.07146955,
      "balance_loss_mlp": 1.03134561,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.7917226182794812,
      "language_loss": 0.77838665,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80117142,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.773571729660034
    },
    {
      "auxiliary_loss_clip": 0.01234043,
      "auxiliary_loss_mlp": 0.01046072,
      "balance_loss_clip": 1.06572568,
      "balance_loss_mlp": 1.03320372,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 2.2790322751487913,
      "language_loss": 0.75887066,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78167182,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.689357042312622
    },
    {
      "auxiliary_loss_clip": 0.01225718,
      "auxiliary_loss_mlp": 0.0090445,
      "balance_loss_clip": 1.06267023,
      "balance_loss_mlp": 1.00006843,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 2.280904582751235,
      "language_loss": 0.80423057,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.82553226,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.772191047668457
    },
    {
      "auxiliary_loss_clip": 0.0121727,
      "auxiliary_loss_mlp": 0.01047339,
      "balance_loss_clip": 1.06569409,
      "balance_loss_mlp": 1.03440535,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.5293441385384687,
      "language_loss": 0.79442298,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81706911,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.6599299907684326
    },
    {
      "auxiliary_loss_clip": 0.01208885,
      "auxiliary_loss_mlp": 0.0103912,
      "balance_loss_clip": 1.06501496,
      "balance_loss_mlp": 1.02570295,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 2.308004425729374,
      "language_loss": 0.58352959,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.6060096,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 4.625114440917969
    },
    {
      "auxiliary_loss_clip": 0.01232401,
      "auxiliary_loss_mlp": 0.0104402,
      "balance_loss_clip": 1.06507993,
      "balance_loss_mlp": 1.03065705,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.7243162341738048,
      "language_loss": 0.85202944,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87479365,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 4.594644546508789
    },
    {
      "auxiliary_loss_clip": 0.01246388,
      "auxiliary_loss_mlp": 0.01042067,
      "balance_loss_clip": 1.06762862,
      "balance_loss_mlp": 1.02932906,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 11.556679632328244,
      "language_loss": 0.71581495,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.7386995,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.5945115089416504
    },
    {
      "auxiliary_loss_clip": 0.01241965,
      "auxiliary_loss_mlp": 0.00905201,
      "balance_loss_clip": 1.06988263,
      "balance_loss_mlp": 1.00010514,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.651563606905304,
      "language_loss": 0.82601202,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84748363,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.661402463912964
    },
    {
      "auxiliary_loss_clip": 0.0123279,
      "auxiliary_loss_mlp": 0.01050011,
      "balance_loss_clip": 1.07187605,
      "balance_loss_mlp": 1.03675449,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.392928301527763,
      "language_loss": 0.86745232,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.89028037,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.677765369415283
    },
    {
      "auxiliary_loss_clip": 0.01255358,
      "auxiliary_loss_mlp": 0.0104555,
      "balance_loss_clip": 1.0663991,
      "balance_loss_mlp": 1.03238297,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.674528010537765,
      "language_loss": 0.76170087,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78470993,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 2.605517625808716
    },
    {
      "auxiliary_loss_clip": 0.01188651,
      "auxiliary_loss_mlp": 0.01047375,
      "balance_loss_clip": 1.05935884,
      "balance_loss_mlp": 1.03545451,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 2.0265530670892438,
      "language_loss": 0.91418999,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93655026,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 2.7273459434509277
    },
    {
      "auxiliary_loss_clip": 0.01194685,
      "auxiliary_loss_mlp": 0.01045196,
      "balance_loss_clip": 1.05591083,
      "balance_loss_mlp": 1.03284025,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 2.0573348840441725,
      "language_loss": 0.82745475,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.84985358,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.7862963676452637
    },
    {
      "auxiliary_loss_clip": 0.0122529,
      "auxiliary_loss_mlp": 0.01049722,
      "balance_loss_clip": 1.06148589,
      "balance_loss_mlp": 1.03660345,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.1480099314361807,
      "language_loss": 0.78681397,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.80956417,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.8317534923553467
    },
    {
      "auxiliary_loss_clip": 0.01230031,
      "auxiliary_loss_mlp": 0.01042606,
      "balance_loss_clip": 1.06885004,
      "balance_loss_mlp": 1.02992177,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.207214720602692,
      "language_loss": 0.77226299,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79498935,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.6664493083953857
    },
    {
      "auxiliary_loss_clip": 0.01218399,
      "auxiliary_loss_mlp": 0.01047872,
      "balance_loss_clip": 1.06662893,
      "balance_loss_mlp": 1.0339185,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 5.234909730932975,
      "language_loss": 0.77926636,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.80192906,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.757124185562134
    },
    {
      "auxiliary_loss_clip": 0.0124359,
      "auxiliary_loss_mlp": 0.01041862,
      "balance_loss_clip": 1.06897163,
      "balance_loss_mlp": 1.02924931,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.9094891303501769,
      "language_loss": 0.81302583,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83588034,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.626836061477661
    },
    {
      "auxiliary_loss_clip": 0.01220655,
      "auxiliary_loss_mlp": 0.01043174,
      "balance_loss_clip": 1.06531978,
      "balance_loss_mlp": 1.0310564,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 2.0152908931436992,
      "language_loss": 0.84769416,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.87033248,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.745676040649414
    },
    {
      "auxiliary_loss_clip": 0.01241478,
      "auxiliary_loss_mlp": 0.01034599,
      "balance_loss_clip": 1.06855321,
      "balance_loss_mlp": 1.02142072,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 2.066597440399345,
      "language_loss": 0.86752355,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.8902843,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.6355204582214355
    },
    {
      "auxiliary_loss_clip": 0.01105962,
      "auxiliary_loss_mlp": 0.01007038,
      "balance_loss_clip": 1.03579068,
      "balance_loss_mlp": 1.0037483,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8036431746686478,
      "language_loss": 0.54482269,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56595272,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.1493844985961914
    },
    {
      "auxiliary_loss_clip": 0.01253625,
      "auxiliary_loss_mlp": 0.0104091,
      "balance_loss_clip": 1.07194495,
      "balance_loss_mlp": 1.02819645,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.5953081243804643,
      "language_loss": 0.78738272,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.81032807,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.718031644821167
    },
    {
      "auxiliary_loss_clip": 0.01220622,
      "auxiliary_loss_mlp": 0.01048233,
      "balance_loss_clip": 1.06321311,
      "balance_loss_mlp": 1.03508997,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.194277907543565,
      "language_loss": 0.75274909,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77543759,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.716160774230957
    },
    {
      "auxiliary_loss_clip": 0.01193536,
      "auxiliary_loss_mlp": 0.01056046,
      "balance_loss_clip": 1.0590806,
      "balance_loss_mlp": 1.0421288,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.0222087729959295,
      "language_loss": 0.8345139,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85700977,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.699035882949829
    },
    {
      "auxiliary_loss_clip": 0.0125085,
      "auxiliary_loss_mlp": 0.0103513,
      "balance_loss_clip": 1.06874585,
      "balance_loss_mlp": 1.02280974,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 2.6872436491135345,
      "language_loss": 1.05909801,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08195782,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.6620290279388428
    },
    {
      "auxiliary_loss_clip": 0.0123915,
      "auxiliary_loss_mlp": 0.0104906,
      "balance_loss_clip": 1.06948686,
      "balance_loss_mlp": 1.03727055,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 6.705497596123911,
      "language_loss": 0.80420691,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82708895,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.680176019668579
    },
    {
      "auxiliary_loss_clip": 0.01208859,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.06364906,
      "balance_loss_mlp": 1.02524066,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.341163217197182,
      "language_loss": 0.76308239,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78555405,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.6780452728271484
    },
    {
      "auxiliary_loss_clip": 0.01237417,
      "auxiliary_loss_mlp": 0.01046459,
      "balance_loss_clip": 1.06573772,
      "balance_loss_mlp": 1.03462744,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 1.869843524960701,
      "language_loss": 0.72059029,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74342906,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.702254056930542
    },
    {
      "auxiliary_loss_clip": 0.01253129,
      "auxiliary_loss_mlp": 0.01053224,
      "balance_loss_clip": 1.07109606,
      "balance_loss_mlp": 1.0397594,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.1275367761564508,
      "language_loss": 0.75355172,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77661526,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.584956407546997
    },
    {
      "auxiliary_loss_clip": 0.01250039,
      "auxiliary_loss_mlp": 0.01052737,
      "balance_loss_clip": 1.06954885,
      "balance_loss_mlp": 1.04092312,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 2.901502749854504,
      "language_loss": 0.71115017,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73417795,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.635348081588745
    },
    {
      "auxiliary_loss_clip": 0.01223501,
      "auxiliary_loss_mlp": 0.01044684,
      "balance_loss_clip": 1.06603074,
      "balance_loss_mlp": 1.03204203,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 2.2292796938965296,
      "language_loss": 0.79692096,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81960285,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.7036373615264893
    },
    {
      "auxiliary_loss_clip": 0.01248995,
      "auxiliary_loss_mlp": 0.01048845,
      "balance_loss_clip": 1.06862462,
      "balance_loss_mlp": 1.03668547,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.5716066446742936,
      "language_loss": 0.90122592,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92420435,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.55519962310791
    },
    {
      "auxiliary_loss_clip": 0.01243929,
      "auxiliary_loss_mlp": 0.01038572,
      "balance_loss_clip": 1.06935239,
      "balance_loss_mlp": 1.02607882,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 2.0020157754479238,
      "language_loss": 0.81563252,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83845758,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 3.5288023948669434
    },
    {
      "auxiliary_loss_clip": 0.01229936,
      "auxiliary_loss_mlp": 0.01049107,
      "balance_loss_clip": 1.06806111,
      "balance_loss_mlp": 1.03631592,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.153704822452279,
      "language_loss": 0.83123863,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85402906,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 3.6069159507751465
    },
    {
      "auxiliary_loss_clip": 0.01094908,
      "auxiliary_loss_mlp": 0.01009352,
      "balance_loss_clip": 1.01770949,
      "balance_loss_mlp": 1.00608599,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0496138774278911,
      "language_loss": 0.70145881,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.7225014,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 4.880287170410156
    },
    {
      "auxiliary_loss_clip": 0.01251519,
      "auxiliary_loss_mlp": 0.0104943,
      "balance_loss_clip": 1.07258582,
      "balance_loss_mlp": 1.03797388,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.7161723199644814,
      "language_loss": 0.77571851,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79872799,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.588991403579712
    },
    {
      "auxiliary_loss_clip": 0.01221987,
      "auxiliary_loss_mlp": 0.01036802,
      "balance_loss_clip": 1.06430507,
      "balance_loss_mlp": 1.02365339,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.081442943514841,
      "language_loss": 0.83486724,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.85745513,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 2.5881857872009277
    },
    {
      "auxiliary_loss_clip": 0.01237545,
      "auxiliary_loss_mlp": 0.0104704,
      "balance_loss_clip": 1.06524289,
      "balance_loss_mlp": 1.03519118,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 2.1560034193387763,
      "language_loss": 0.72104478,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.74389064,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.6237778663635254
    },
    {
      "auxiliary_loss_clip": 0.01230767,
      "auxiliary_loss_mlp": 0.01042277,
      "balance_loss_clip": 1.06772232,
      "balance_loss_mlp": 1.02997458,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 2.2375713189923787,
      "language_loss": 0.79863828,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82136869,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 2.6641170978546143
    },
    {
      "auxiliary_loss_clip": 0.012426,
      "auxiliary_loss_mlp": 0.01050887,
      "balance_loss_clip": 1.0683589,
      "balance_loss_mlp": 1.03788733,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 2.095338919644111,
      "language_loss": 0.76424563,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.78718048,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.637392044067383
    },
    {
      "auxiliary_loss_clip": 0.01233871,
      "auxiliary_loss_mlp": 0.01046302,
      "balance_loss_clip": 1.06808162,
      "balance_loss_mlp": 1.03350496,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.5814868522524108,
      "language_loss": 0.79662538,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81942713,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.7038187980651855
    },
    {
      "auxiliary_loss_clip": 0.01223853,
      "auxiliary_loss_mlp": 0.01039662,
      "balance_loss_clip": 1.06073618,
      "balance_loss_mlp": 1.02649498,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 3.2435701265919237,
      "language_loss": 0.71751755,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.74015272,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.76155686378479
    },
    {
      "auxiliary_loss_clip": 0.01247624,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.06815863,
      "balance_loss_mlp": 1.02138734,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 1.8827620668985279,
      "language_loss": 0.78564346,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80845153,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.609516143798828
    },
    {
      "auxiliary_loss_clip": 0.01249516,
      "auxiliary_loss_mlp": 0.00904732,
      "balance_loss_clip": 1.06966043,
      "balance_loss_mlp": 1.00019479,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.034003764693577,
      "language_loss": 0.88127238,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90281487,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.6274056434631348
    },
    {
      "auxiliary_loss_clip": 0.01230322,
      "auxiliary_loss_mlp": 0.01043124,
      "balance_loss_clip": 1.06662238,
      "balance_loss_mlp": 1.03042173,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 3.499306817542677,
      "language_loss": 0.80381364,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.8265481,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.7294859886169434
    },
    {
      "auxiliary_loss_clip": 0.01238227,
      "auxiliary_loss_mlp": 0.01048136,
      "balance_loss_clip": 1.06519175,
      "balance_loss_mlp": 1.03656101,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.837421589922874,
      "language_loss": 0.75560164,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77846527,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.6592860221862793
    },
    {
      "auxiliary_loss_clip": 0.01247041,
      "auxiliary_loss_mlp": 0.01042285,
      "balance_loss_clip": 1.06783283,
      "balance_loss_mlp": 1.02976227,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.7470430009480657,
      "language_loss": 0.84312403,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86601722,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.548546314239502
    },
    {
      "auxiliary_loss_clip": 0.01216805,
      "auxiliary_loss_mlp": 0.01037687,
      "balance_loss_clip": 1.06045675,
      "balance_loss_mlp": 1.02440715,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.2238371917358237,
      "language_loss": 0.86623329,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88877821,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.628115653991699
    },
    {
      "auxiliary_loss_clip": 0.01246884,
      "auxiliary_loss_mlp": 0.00905757,
      "balance_loss_clip": 1.06709838,
      "balance_loss_mlp": 1.00026035,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.1847477466896783,
      "language_loss": 0.87533277,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89685917,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.5933573246002197
    },
    {
      "auxiliary_loss_clip": 0.01233833,
      "auxiliary_loss_mlp": 0.01042075,
      "balance_loss_clip": 1.06652391,
      "balance_loss_mlp": 1.02996874,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 1.8696162376276233,
      "language_loss": 0.76996046,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.7927196,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.5980277061462402
    },
    {
      "auxiliary_loss_clip": 0.01246983,
      "auxiliary_loss_mlp": 0.01040499,
      "balance_loss_clip": 1.06960595,
      "balance_loss_mlp": 1.02863145,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 2.2891667874376758,
      "language_loss": 0.85777211,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88064688,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.5917611122131348
    },
    {
      "auxiliary_loss_clip": 0.01229512,
      "auxiliary_loss_mlp": 0.01044324,
      "balance_loss_clip": 1.06921637,
      "balance_loss_mlp": 1.03280222,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.465729579387475,
      "language_loss": 0.79667562,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.81941396,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.6555187702178955
    },
    {
      "auxiliary_loss_clip": 0.01235639,
      "auxiliary_loss_mlp": 0.01045217,
      "balance_loss_clip": 1.06506419,
      "balance_loss_mlp": 1.03255653,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.5801537133541337,
      "language_loss": 0.8806296,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90343809,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.6427667140960693
    },
    {
      "auxiliary_loss_clip": 0.01120856,
      "auxiliary_loss_mlp": 0.01009578,
      "balance_loss_clip": 1.02299595,
      "balance_loss_mlp": 1.00562024,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9323055309936488,
      "language_loss": 0.64423859,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66554296,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.029900550842285
    },
    {
      "auxiliary_loss_clip": 0.01241314,
      "auxiliary_loss_mlp": 0.01040582,
      "balance_loss_clip": 1.06608129,
      "balance_loss_mlp": 1.02738535,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 7.672495099742778,
      "language_loss": 0.80744618,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.8302651,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.643437623977661
    },
    {
      "auxiliary_loss_clip": 0.01236998,
      "auxiliary_loss_mlp": 0.00904671,
      "balance_loss_clip": 1.06869709,
      "balance_loss_mlp": 1.00027514,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 2.5869208336683713,
      "language_loss": 0.81488645,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.83630311,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.715752124786377
    },
    {
      "auxiliary_loss_clip": 0.0123688,
      "auxiliary_loss_mlp": 0.0105117,
      "balance_loss_clip": 1.06591415,
      "balance_loss_mlp": 1.03896856,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.435409854155425,
      "language_loss": 0.86134261,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.8842231,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.6805648803710938
    },
    {
      "auxiliary_loss_clip": 0.01229765,
      "auxiliary_loss_mlp": 0.01040711,
      "balance_loss_clip": 1.06374431,
      "balance_loss_mlp": 1.02808022,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 3.063596096880423,
      "language_loss": 0.91094059,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93364537,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.5761678218841553
    },
    {
      "auxiliary_loss_clip": 0.01241466,
      "auxiliary_loss_mlp": 0.01038266,
      "balance_loss_clip": 1.07165492,
      "balance_loss_mlp": 1.02666688,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.6970665066365427,
      "language_loss": 0.78795844,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81075579,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.6410744190216064
    },
    {
      "auxiliary_loss_clip": 0.01244953,
      "auxiliary_loss_mlp": 0.01044344,
      "balance_loss_clip": 1.06876123,
      "balance_loss_mlp": 1.0336802,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 4.247073861367485,
      "language_loss": 0.77238631,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79527926,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 4.426119327545166
    },
    {
      "auxiliary_loss_clip": 0.01216804,
      "auxiliary_loss_mlp": 0.01041646,
      "balance_loss_clip": 1.06038451,
      "balance_loss_mlp": 1.0290513,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 2.00017147990503,
      "language_loss": 0.92246521,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.9450497,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 3.6315696239471436
    },
    {
      "auxiliary_loss_clip": 0.01251056,
      "auxiliary_loss_mlp": 0.01047094,
      "balance_loss_clip": 1.07127094,
      "balance_loss_mlp": 1.03576362,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.149310838771954,
      "language_loss": 0.83156812,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85454971,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 3.5422534942626953
    },
    {
      "auxiliary_loss_clip": 0.01215293,
      "auxiliary_loss_mlp": 0.0103288,
      "balance_loss_clip": 1.06103086,
      "balance_loss_mlp": 1.02182937,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.8543751400027295,
      "language_loss": 0.88252008,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90500182,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.661184549331665
    },
    {
      "auxiliary_loss_clip": 0.01082551,
      "auxiliary_loss_mlp": 0.01007984,
      "balance_loss_clip": 1.01810169,
      "balance_loss_mlp": 1.00459886,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9304462473173294,
      "language_loss": 0.60439789,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62530327,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 3.2278647422790527
    },
    {
      "auxiliary_loss_clip": 0.01212548,
      "auxiliary_loss_mlp": 0.01048946,
      "balance_loss_clip": 1.06145358,
      "balance_loss_mlp": 1.0363456,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.9896001304066586,
      "language_loss": 0.82682079,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.84943575,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.0539743900299072
    },
    {
      "auxiliary_loss_clip": 0.01198213,
      "auxiliary_loss_mlp": 0.00904343,
      "balance_loss_clip": 1.05953288,
      "balance_loss_mlp": 1.0003283,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.8029351341042543,
      "language_loss": 0.84020317,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.86122876,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.8490402698516846
    },
    {
      "auxiliary_loss_clip": 0.01216047,
      "auxiliary_loss_mlp": 0.01036471,
      "balance_loss_clip": 1.06101274,
      "balance_loss_mlp": 1.02535474,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.8310603587796621,
      "language_loss": 0.77488869,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79741389,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.6615865230560303
    },
    {
      "auxiliary_loss_clip": 0.01230511,
      "auxiliary_loss_mlp": 0.01041165,
      "balance_loss_clip": 1.06411803,
      "balance_loss_mlp": 1.02829087,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.6466175266369905,
      "language_loss": 0.82261348,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84533024,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.766314744949341
    },
    {
      "auxiliary_loss_clip": 0.01233867,
      "auxiliary_loss_mlp": 0.01041718,
      "balance_loss_clip": 1.06872606,
      "balance_loss_mlp": 1.03091216,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 3.3602691705873977,
      "language_loss": 0.75983632,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78259218,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.6196389198303223
    },
    {
      "auxiliary_loss_clip": 0.01224944,
      "auxiliary_loss_mlp": 0.00903724,
      "balance_loss_clip": 1.06091237,
      "balance_loss_mlp": 1.00039101,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.3202463136366163,
      "language_loss": 0.73204088,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75332749,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.6319210529327393
    },
    {
      "auxiliary_loss_clip": 0.01236913,
      "auxiliary_loss_mlp": 0.01043901,
      "balance_loss_clip": 1.06371951,
      "balance_loss_mlp": 1.03250504,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.8917938180205203,
      "language_loss": 0.78826463,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.81107277,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.9556021690368652
    },
    {
      "auxiliary_loss_clip": 0.01239619,
      "auxiliary_loss_mlp": 0.01039201,
      "balance_loss_clip": 1.06793332,
      "balance_loss_mlp": 1.02694631,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 5.887512352997942,
      "language_loss": 0.70216918,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72495747,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.627758502960205
    },
    {
      "auxiliary_loss_clip": 0.01222372,
      "auxiliary_loss_mlp": 0.0104203,
      "balance_loss_clip": 1.06096363,
      "balance_loss_mlp": 1.03037179,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 1.9322193861735835,
      "language_loss": 0.90868974,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93133378,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.705861806869507
    },
    {
      "auxiliary_loss_clip": 0.01228346,
      "auxiliary_loss_mlp": 0.01041602,
      "balance_loss_clip": 1.06596088,
      "balance_loss_mlp": 1.02986598,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.9110709300296072,
      "language_loss": 0.83121812,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.8539176,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.7258377075195312
    },
    {
      "auxiliary_loss_clip": 0.01217345,
      "auxiliary_loss_mlp": 0.01036838,
      "balance_loss_clip": 1.06364822,
      "balance_loss_mlp": 1.02576923,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.0360150351790325,
      "language_loss": 0.81269228,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83523405,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.6522481441497803
    },
    {
      "auxiliary_loss_clip": 0.01183492,
      "auxiliary_loss_mlp": 0.01039077,
      "balance_loss_clip": 1.05125511,
      "balance_loss_mlp": 1.02732909,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.712858487240803,
      "language_loss": 0.75755453,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.77978021,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.676504611968994
    },
    {
      "auxiliary_loss_clip": 0.01222646,
      "auxiliary_loss_mlp": 0.01034199,
      "balance_loss_clip": 1.06306052,
      "balance_loss_mlp": 1.02373874,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 2.292919696286387,
      "language_loss": 0.85786366,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.88043207,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.6288504600524902
    },
    {
      "auxiliary_loss_clip": 0.01218904,
      "auxiliary_loss_mlp": 0.01039104,
      "balance_loss_clip": 1.06031132,
      "balance_loss_mlp": 1.02780914,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 1.9490839685400023,
      "language_loss": 0.78716999,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80975008,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.69368577003479
    },
    {
      "auxiliary_loss_clip": 0.01199802,
      "auxiliary_loss_mlp": 0.01041332,
      "balance_loss_clip": 1.05221546,
      "balance_loss_mlp": 1.02985275,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.4416835065114975,
      "language_loss": 0.87239933,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89481074,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.742976665496826
    },
    {
      "auxiliary_loss_clip": 0.01245466,
      "auxiliary_loss_mlp": 0.01037746,
      "balance_loss_clip": 1.06888521,
      "balance_loss_mlp": 1.02640319,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.416737997793095,
      "language_loss": 0.82338881,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84622097,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.599801778793335
    },
    {
      "auxiliary_loss_clip": 0.01249708,
      "auxiliary_loss_mlp": 0.01037646,
      "balance_loss_clip": 1.06772816,
      "balance_loss_mlp": 1.02501535,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.3312047985383737,
      "language_loss": 0.70418489,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72705841,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.7779123783111572
    },
    {
      "auxiliary_loss_clip": 0.0121644,
      "auxiliary_loss_mlp": 0.01041662,
      "balance_loss_clip": 1.06400287,
      "balance_loss_mlp": 1.03008664,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.6389573272406377,
      "language_loss": 0.78370261,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80628359,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.7498652935028076
    },
    {
      "auxiliary_loss_clip": 0.01208659,
      "auxiliary_loss_mlp": 0.01047873,
      "balance_loss_clip": 1.05864358,
      "balance_loss_mlp": 1.03644681,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.2440406763505196,
      "language_loss": 0.85606825,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.87863362,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.688096046447754
    },
    {
      "auxiliary_loss_clip": 0.0120296,
      "auxiliary_loss_mlp": 0.01042662,
      "balance_loss_clip": 1.06112742,
      "balance_loss_mlp": 1.02968633,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 4.992258709069902,
      "language_loss": 0.89089179,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91334796,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.6903276443481445
    },
    {
      "auxiliary_loss_clip": 0.01200865,
      "auxiliary_loss_mlp": 0.010426,
      "balance_loss_clip": 1.06263876,
      "balance_loss_mlp": 1.0322113,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.8423001382516786,
      "language_loss": 0.82228285,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.8447175,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.7220356464385986
    },
    {
      "auxiliary_loss_clip": 0.01236005,
      "auxiliary_loss_mlp": 0.01042807,
      "balance_loss_clip": 1.06702256,
      "balance_loss_mlp": 1.03045154,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.663701061348492,
      "language_loss": 0.82204258,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84483063,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.5878210067749023
    },
    {
      "auxiliary_loss_clip": 0.01233324,
      "auxiliary_loss_mlp": 0.00904531,
      "balance_loss_clip": 1.06765187,
      "balance_loss_mlp": 1.00043845,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.1861804651230163,
      "language_loss": 0.79725325,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.81863183,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 4.511429309844971
    },
    {
      "auxiliary_loss_clip": 0.01236786,
      "auxiliary_loss_mlp": 0.01038087,
      "balance_loss_clip": 1.06583047,
      "balance_loss_mlp": 1.02704215,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 2.16426998038362,
      "language_loss": 0.83986217,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86261082,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 3.710207462310791
    },
    {
      "auxiliary_loss_clip": 0.01204769,
      "auxiliary_loss_mlp": 0.01038627,
      "balance_loss_clip": 1.0601064,
      "balance_loss_mlp": 1.02711177,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 2.8386710985636316,
      "language_loss": 0.83812082,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86055475,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 3.6798107624053955
    },
    {
      "auxiliary_loss_clip": 0.01220527,
      "auxiliary_loss_mlp": 0.01039598,
      "balance_loss_clip": 1.06085753,
      "balance_loss_mlp": 1.02908373,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.768393507840088,
      "language_loss": 0.86846256,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89106381,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 2.682724714279175
    },
    {
      "auxiliary_loss_clip": 0.01232581,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.06285453,
      "balance_loss_mlp": 1.02697682,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.742672924354988,
      "language_loss": 0.86329097,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88600194,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 2.6389522552490234
    },
    {
      "auxiliary_loss_clip": 0.01174725,
      "auxiliary_loss_mlp": 0.01039919,
      "balance_loss_clip": 1.05456662,
      "balance_loss_mlp": 1.02818942,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 1.8805981308108186,
      "language_loss": 0.81691331,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.83905977,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.7487032413482666
    },
    {
      "auxiliary_loss_clip": 0.01228863,
      "auxiliary_loss_mlp": 0.01043461,
      "balance_loss_clip": 1.06776118,
      "balance_loss_mlp": 1.03087795,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.959612368159302,
      "language_loss": 0.92613262,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94885588,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.6631734371185303
    },
    {
      "auxiliary_loss_clip": 0.0119159,
      "auxiliary_loss_mlp": 0.01038038,
      "balance_loss_clip": 1.05872631,
      "balance_loss_mlp": 1.02594411,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 3.4143315690517864,
      "language_loss": 0.85512745,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87742376,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.709925651550293
    },
    {
      "auxiliary_loss_clip": 0.01192187,
      "auxiliary_loss_mlp": 0.01047967,
      "balance_loss_clip": 1.05728137,
      "balance_loss_mlp": 1.03521156,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 4.584912517175614,
      "language_loss": 0.91424859,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93665016,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.6827890872955322
    },
    {
      "auxiliary_loss_clip": 0.01205825,
      "auxiliary_loss_mlp": 0.00904528,
      "balance_loss_clip": 1.05665636,
      "balance_loss_mlp": 1.00042248,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 2.0390619753989307,
      "language_loss": 0.83878678,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.85989022,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.7286574840545654
    },
    {
      "auxiliary_loss_clip": 0.01230795,
      "auxiliary_loss_mlp": 0.01041406,
      "balance_loss_clip": 1.07151794,
      "balance_loss_mlp": 1.02911568,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.2015420786469986,
      "language_loss": 0.82618928,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84891123,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.6499552726745605
    },
    {
      "auxiliary_loss_clip": 0.01221364,
      "auxiliary_loss_mlp": 0.01039908,
      "balance_loss_clip": 1.0654707,
      "balance_loss_mlp": 1.02743244,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.368878150032641,
      "language_loss": 0.84886128,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87147397,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.697488307952881
    },
    {
      "auxiliary_loss_clip": 0.01191213,
      "auxiliary_loss_mlp": 0.01037172,
      "balance_loss_clip": 1.05669856,
      "balance_loss_mlp": 1.02537024,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 2.9692153307408424,
      "language_loss": 0.81923807,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.84152186,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.696308135986328
    },
    {
      "auxiliary_loss_clip": 0.01220959,
      "auxiliary_loss_mlp": 0.01046586,
      "balance_loss_clip": 1.05863214,
      "balance_loss_mlp": 1.033705,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 2.1307756509622666,
      "language_loss": 0.83819121,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.86086667,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.799945116043091
    },
    {
      "auxiliary_loss_clip": 0.01242033,
      "auxiliary_loss_mlp": 0.01037683,
      "balance_loss_clip": 1.06875074,
      "balance_loss_mlp": 1.02536261,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 3.793218360158413,
      "language_loss": 0.876715,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89951217,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.605882406234741
    },
    {
      "auxiliary_loss_clip": 0.01154775,
      "auxiliary_loss_mlp": 0.01041899,
      "balance_loss_clip": 1.05295491,
      "balance_loss_mlp": 1.02962637,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 1.7484354833061715,
      "language_loss": 0.8911283,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.913095,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.9091031551361084
    },
    {
      "auxiliary_loss_clip": 0.01196524,
      "auxiliary_loss_mlp": 0.01050185,
      "balance_loss_clip": 1.05774462,
      "balance_loss_mlp": 1.03695297,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.8887576114826086,
      "language_loss": 0.78772068,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81018782,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.804184675216675
    },
    {
      "auxiliary_loss_clip": 0.01222836,
      "auxiliary_loss_mlp": 0.01042196,
      "balance_loss_clip": 1.06381416,
      "balance_loss_mlp": 1.03119254,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.2303173486502232,
      "language_loss": 0.86662412,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.88927448,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.6483800411224365
    },
    {
      "auxiliary_loss_clip": 0.01243879,
      "auxiliary_loss_mlp": 0.01045536,
      "balance_loss_clip": 1.07075644,
      "balance_loss_mlp": 1.03260136,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.058572366889061,
      "language_loss": 0.84566295,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.8685571,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.654818534851074
    },
    {
      "auxiliary_loss_clip": 0.01243625,
      "auxiliary_loss_mlp": 0.01040146,
      "balance_loss_clip": 1.06797111,
      "balance_loss_mlp": 1.02939343,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.6304351938359976,
      "language_loss": 0.8003341,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82317185,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.609497308731079
    },
    {
      "auxiliary_loss_clip": 0.01221744,
      "auxiliary_loss_mlp": 0.01038949,
      "balance_loss_clip": 1.06525207,
      "balance_loss_mlp": 1.02726674,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 2.9423051515396104,
      "language_loss": 0.64935958,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67196655,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.6849520206451416
    },
    {
      "auxiliary_loss_clip": 0.012138,
      "auxiliary_loss_mlp": 0.01042588,
      "balance_loss_clip": 1.06184459,
      "balance_loss_mlp": 1.03017247,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.333491188209354,
      "language_loss": 0.74928808,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77185196,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.694918632507324
    },
    {
      "auxiliary_loss_clip": 0.01225165,
      "auxiliary_loss_mlp": 0.01044442,
      "balance_loss_clip": 1.06616092,
      "balance_loss_mlp": 1.03148985,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.080848387992286,
      "language_loss": 0.7609486,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.78364462,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.630023956298828
    },
    {
      "auxiliary_loss_clip": 0.01227052,
      "auxiliary_loss_mlp": 0.00904762,
      "balance_loss_clip": 1.06416225,
      "balance_loss_mlp": 1.00059962,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 2.3924358842649758,
      "language_loss": 0.76408863,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78540683,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.7115299701690674
    },
    {
      "auxiliary_loss_clip": 0.01215207,
      "auxiliary_loss_mlp": 0.01040881,
      "balance_loss_clip": 1.06180263,
      "balance_loss_mlp": 1.03015268,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 3.1192203110835472,
      "language_loss": 0.79547989,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81804085,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.6610946655273438
    },
    {
      "auxiliary_loss_clip": 0.01224188,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.06446826,
      "balance_loss_mlp": 1.02094448,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 2.46497054407943,
      "language_loss": 0.75129509,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77385741,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.657320261001587
    },
    {
      "auxiliary_loss_clip": 0.01214106,
      "auxiliary_loss_mlp": 0.01051707,
      "balance_loss_clip": 1.06471705,
      "balance_loss_mlp": 1.04101408,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.9216766292552048,
      "language_loss": 0.82822073,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.85087889,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 3.6183528900146484
    },
    {
      "auxiliary_loss_clip": 0.0124958,
      "auxiliary_loss_mlp": 0.01037484,
      "balance_loss_clip": 1.06891942,
      "balance_loss_mlp": 1.02442431,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.172226335581956,
      "language_loss": 0.89804798,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92091858,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 3.5813164710998535
    },
    {
      "auxiliary_loss_clip": 0.01211046,
      "auxiliary_loss_mlp": 0.0103803,
      "balance_loss_clip": 1.06193113,
      "balance_loss_mlp": 1.0259366,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.2008299996610643,
      "language_loss": 0.66230786,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.6847986,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 4.597840785980225
    },
    {
      "auxiliary_loss_clip": 0.01225008,
      "auxiliary_loss_mlp": 0.01038478,
      "balance_loss_clip": 1.06877279,
      "balance_loss_mlp": 1.02681911,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 1.8850143607135659,
      "language_loss": 0.88019598,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.90283084,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 2.6702113151550293
    },
    {
      "auxiliary_loss_clip": 0.01239877,
      "auxiliary_loss_mlp": 0.01042497,
      "balance_loss_clip": 1.07094049,
      "balance_loss_mlp": 1.03070748,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 2.3856077175327326,
      "language_loss": 0.72045195,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74327576,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.675217866897583
    },
    {
      "auxiliary_loss_clip": 0.01249126,
      "auxiliary_loss_mlp": 0.01042234,
      "balance_loss_clip": 1.06960595,
      "balance_loss_mlp": 1.0301398,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.6997760179103816,
      "language_loss": 0.83869249,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86160612,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.574828624725342
    },
    {
      "auxiliary_loss_clip": 0.01223543,
      "auxiliary_loss_mlp": 0.01035051,
      "balance_loss_clip": 1.06425953,
      "balance_loss_mlp": 1.02472115,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 2.8789758097832094,
      "language_loss": 0.71229672,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73488259,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.7508442401885986
    },
    {
      "auxiliary_loss_clip": 0.01233053,
      "auxiliary_loss_mlp": 0.01038092,
      "balance_loss_clip": 1.06985545,
      "balance_loss_mlp": 1.02745831,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 2.014727806153828,
      "language_loss": 0.77367127,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79638273,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.614114999771118
    },
    {
      "auxiliary_loss_clip": 0.01223793,
      "auxiliary_loss_mlp": 0.01047388,
      "balance_loss_clip": 1.06437409,
      "balance_loss_mlp": 1.03547907,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 2.0577886078637824,
      "language_loss": 0.94537956,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96809137,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.67702317237854
    },
    {
      "auxiliary_loss_clip": 0.01177858,
      "auxiliary_loss_mlp": 0.01037531,
      "balance_loss_clip": 1.05111313,
      "balance_loss_mlp": 1.02530003,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 2.672295321256221,
      "language_loss": 0.75241303,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77456695,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.797870635986328
    },
    {
      "auxiliary_loss_clip": 0.01198433,
      "auxiliary_loss_mlp": 0.01039751,
      "balance_loss_clip": 1.05740571,
      "balance_loss_mlp": 1.02933192,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.3118893109056917,
      "language_loss": 0.75535548,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77773732,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.7493226528167725
    },
    {
      "auxiliary_loss_clip": 0.01213145,
      "auxiliary_loss_mlp": 0.01049132,
      "balance_loss_clip": 1.06286085,
      "balance_loss_mlp": 1.0371635,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.392873363930299,
      "language_loss": 0.81939316,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.84201598,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.743086099624634
    },
    {
      "auxiliary_loss_clip": 0.01215973,
      "auxiliary_loss_mlp": 0.01035211,
      "balance_loss_clip": 1.06632686,
      "balance_loss_mlp": 1.02452374,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.269236825463008,
      "language_loss": 0.88620746,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90871918,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.6927576065063477
    },
    {
      "auxiliary_loss_clip": 0.01221756,
      "auxiliary_loss_mlp": 0.01042215,
      "balance_loss_clip": 1.06578124,
      "balance_loss_mlp": 1.02990115,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 2.086725427315688,
      "language_loss": 0.76954955,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79218924,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.66229510307312
    },
    {
      "auxiliary_loss_clip": 0.01216129,
      "auxiliary_loss_mlp": 0.0104466,
      "balance_loss_clip": 1.06658077,
      "balance_loss_mlp": 1.03208971,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.4992979724500533,
      "language_loss": 0.88746262,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.91007048,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.7037131786346436
    },
    {
      "auxiliary_loss_clip": 0.01227965,
      "auxiliary_loss_mlp": 0.01040211,
      "balance_loss_clip": 1.06404626,
      "balance_loss_mlp": 1.02880228,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.8549567827389946,
      "language_loss": 0.71785808,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74053985,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.6640844345092773
    },
    {
      "auxiliary_loss_clip": 0.01192312,
      "auxiliary_loss_mlp": 0.0104213,
      "balance_loss_clip": 1.0604142,
      "balance_loss_mlp": 1.03102565,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.7985117535643842,
      "language_loss": 0.78651655,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.80886102,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.820936918258667
    },
    {
      "auxiliary_loss_clip": 0.01248709,
      "auxiliary_loss_mlp": 0.0103942,
      "balance_loss_clip": 1.06960225,
      "balance_loss_mlp": 1.0280298,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 2.4495165147736064,
      "language_loss": 0.82122386,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84410512,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.6125829219818115
    },
    {
      "auxiliary_loss_clip": 0.01212643,
      "auxiliary_loss_mlp": 0.01047717,
      "balance_loss_clip": 1.06282079,
      "balance_loss_mlp": 1.03591502,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.5043876429230028,
      "language_loss": 0.92613089,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94873446,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.589855194091797
    },
    {
      "auxiliary_loss_clip": 0.01228871,
      "auxiliary_loss_mlp": 0.0103821,
      "balance_loss_clip": 1.06440961,
      "balance_loss_mlp": 1.02631259,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.5376101224513334,
      "language_loss": 0.8316685,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.85433936,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.598825454711914
    },
    {
      "auxiliary_loss_clip": 0.01236142,
      "auxiliary_loss_mlp": 0.0104217,
      "balance_loss_clip": 1.06821489,
      "balance_loss_mlp": 1.03028512,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.5633934081993424,
      "language_loss": 0.70453846,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72732162,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.868490695953369
    },
    {
      "auxiliary_loss_clip": 0.01208598,
      "auxiliary_loss_mlp": 0.00904455,
      "balance_loss_clip": 1.05861592,
      "balance_loss_mlp": 1.00084937,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 2.576300561713841,
      "language_loss": 0.79224831,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81337887,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.8319170475006104
    },
    {
      "auxiliary_loss_clip": 0.01214632,
      "auxiliary_loss_mlp": 0.01038223,
      "balance_loss_clip": 1.06457734,
      "balance_loss_mlp": 1.02600384,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.3047051962604272,
      "language_loss": 0.90258467,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.9251132,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.6824612617492676
    },
    {
      "auxiliary_loss_clip": 0.01224069,
      "auxiliary_loss_mlp": 0.0090452,
      "balance_loss_clip": 1.06164396,
      "balance_loss_mlp": 1.00078821,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 2.1277053705588718,
      "language_loss": 0.75981629,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78110218,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.6585769653320312
    },
    {
      "auxiliary_loss_clip": 0.01225182,
      "auxiliary_loss_mlp": 0.01040346,
      "balance_loss_clip": 1.06701446,
      "balance_loss_mlp": 1.02834129,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.2960852097082385,
      "language_loss": 0.72395331,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.74660861,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.6490981578826904
    },
    {
      "auxiliary_loss_clip": 0.01236255,
      "auxiliary_loss_mlp": 0.01044053,
      "balance_loss_clip": 1.0683744,
      "balance_loss_mlp": 1.03155422,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 4.3096157338909,
      "language_loss": 0.88629305,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90909618,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.6158041954040527
    },
    {
      "auxiliary_loss_clip": 0.01120778,
      "auxiliary_loss_mlp": 0.01005351,
      "balance_loss_clip": 1.02597308,
      "balance_loss_mlp": 1.00210822,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9569126937376178,
      "language_loss": 0.64849877,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66975999,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.14691162109375
    },
    {
      "auxiliary_loss_clip": 0.01210428,
      "auxiliary_loss_mlp": 0.01038023,
      "balance_loss_clip": 1.06248403,
      "balance_loss_mlp": 1.02589965,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 3.097025517218932,
      "language_loss": 0.95097983,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97346437,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 4.690893888473511
    },
    {
      "auxiliary_loss_clip": 0.01247755,
      "auxiliary_loss_mlp": 0.0103638,
      "balance_loss_clip": 1.06963658,
      "balance_loss_mlp": 1.02498341,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.437309544686675,
      "language_loss": 0.6619457,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68478703,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 3.461658239364624
    },
    {
      "auxiliary_loss_clip": 0.01232353,
      "auxiliary_loss_mlp": 0.01038814,
      "balance_loss_clip": 1.06618071,
      "balance_loss_mlp": 1.02738225,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.61133696803266,
      "language_loss": 0.78984421,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81255591,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 3.672508955001831
    },
    {
      "auxiliary_loss_clip": 0.01204836,
      "auxiliary_loss_mlp": 0.01039765,
      "balance_loss_clip": 1.06267893,
      "balance_loss_mlp": 1.0282613,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.701588416957624,
      "language_loss": 0.85580891,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87825495,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 2.7581934928894043
    },
    {
      "auxiliary_loss_clip": 0.0124175,
      "auxiliary_loss_mlp": 0.01041479,
      "balance_loss_clip": 1.07441735,
      "balance_loss_mlp": 1.02936101,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 2.466928534426157,
      "language_loss": 0.75764716,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78047943,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.7366509437561035
    },
    {
      "auxiliary_loss_clip": 0.01201414,
      "auxiliary_loss_mlp": 0.01042974,
      "balance_loss_clip": 1.05759025,
      "balance_loss_mlp": 1.03064752,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.6867555385508655,
      "language_loss": 0.72198927,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74443305,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.7229785919189453
    },
    {
      "auxiliary_loss_clip": 0.01220338,
      "auxiliary_loss_mlp": 0.01043082,
      "balance_loss_clip": 1.06244826,
      "balance_loss_mlp": 1.0307914,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.8542906338962375,
      "language_loss": 0.72104073,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74367493,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.627896308898926
    },
    {
      "auxiliary_loss_clip": 0.01234407,
      "auxiliary_loss_mlp": 0.01038541,
      "balance_loss_clip": 1.06620908,
      "balance_loss_mlp": 1.0262866,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.809040260974061,
      "language_loss": 0.88919091,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91192037,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.6485707759857178
    },
    {
      "auxiliary_loss_clip": 0.01222142,
      "auxiliary_loss_mlp": 0.01033549,
      "balance_loss_clip": 1.06380939,
      "balance_loss_mlp": 1.02269459,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 3.16385265339807,
      "language_loss": 0.78412998,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80668694,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.6451315879821777
    },
    {
      "auxiliary_loss_clip": 0.01113986,
      "auxiliary_loss_mlp": 0.01007619,
      "balance_loss_clip": 1.02569592,
      "balance_loss_mlp": 1.0044955,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 1.0009757803174941,
      "language_loss": 0.67629158,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69750762,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.094306230545044
    },
    {
      "auxiliary_loss_clip": 0.01195082,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.06105781,
      "balance_loss_mlp": 1.02265728,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.7584002743795886,
      "language_loss": 0.73733366,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.75962043,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.6853840351104736
    },
    {
      "auxiliary_loss_clip": 0.01233706,
      "auxiliary_loss_mlp": 0.01046152,
      "balance_loss_clip": 1.06899953,
      "balance_loss_mlp": 1.03449333,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.726217441745495,
      "language_loss": 0.88208628,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90488493,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.7004356384277344
    },
    {
      "auxiliary_loss_clip": 0.01240571,
      "auxiliary_loss_mlp": 0.0104401,
      "balance_loss_clip": 1.07362854,
      "balance_loss_mlp": 1.03259599,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 2.2156540859552205,
      "language_loss": 0.78224647,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80509222,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.6583664417266846
    },
    {
      "auxiliary_loss_clip": 0.01232999,
      "auxiliary_loss_mlp": 0.01038759,
      "balance_loss_clip": 1.0668503,
      "balance_loss_mlp": 1.02672517,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 2.533275943630116,
      "language_loss": 0.79463428,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81735182,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.685325860977173
    },
    {
      "auxiliary_loss_clip": 0.01113713,
      "auxiliary_loss_mlp": 0.00894151,
      "balance_loss_clip": 1.02633488,
      "balance_loss_mlp": 1.00010419,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8831709448798463,
      "language_loss": 0.56664395,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.58672261,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.201420545578003
    },
    {
      "auxiliary_loss_clip": 0.01223543,
      "auxiliary_loss_mlp": 0.01041033,
      "balance_loss_clip": 1.0670408,
      "balance_loss_mlp": 1.03028047,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 1.840992867566511,
      "language_loss": 0.81836462,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84101033,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.7404751777648926
    },
    {
      "auxiliary_loss_clip": 0.01225554,
      "auxiliary_loss_mlp": 0.0090384,
      "balance_loss_clip": 1.07028842,
      "balance_loss_mlp": 1.00071597,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 2.2814841238091343,
      "language_loss": 0.80613762,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.82743156,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.789719581604004
    },
    {
      "auxiliary_loss_clip": 0.01201851,
      "auxiliary_loss_mlp": 0.01040197,
      "balance_loss_clip": 1.06283689,
      "balance_loss_mlp": 1.02879429,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 2.8368452341700485,
      "language_loss": 0.8775475,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89996797,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.6956236362457275
    },
    {
      "auxiliary_loss_clip": 0.01221145,
      "auxiliary_loss_mlp": 0.01038401,
      "balance_loss_clip": 1.06130517,
      "balance_loss_mlp": 1.02652216,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.89164742718904,
      "language_loss": 0.86125553,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88385093,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.67973256111145
    },
    {
      "auxiliary_loss_clip": 0.01216974,
      "auxiliary_loss_mlp": 0.01035493,
      "balance_loss_clip": 1.06338787,
      "balance_loss_mlp": 1.02356005,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 2.47894516428094,
      "language_loss": 0.86218107,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88470566,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.6753430366516113
    },
    {
      "auxiliary_loss_clip": 0.01236074,
      "auxiliary_loss_mlp": 0.01058321,
      "balance_loss_clip": 1.06608915,
      "balance_loss_mlp": 1.04647779,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 1.9170683719156407,
      "language_loss": 0.69062966,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71357358,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.7122058868408203
    },
    {
      "auxiliary_loss_clip": 0.01222925,
      "auxiliary_loss_mlp": 0.01046113,
      "balance_loss_clip": 1.06391621,
      "balance_loss_mlp": 1.03478825,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 1.99803282207427,
      "language_loss": 0.79255086,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81524122,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.6956050395965576
    },
    {
      "auxiliary_loss_clip": 0.0123963,
      "auxiliary_loss_mlp": 0.01035048,
      "balance_loss_clip": 1.06659651,
      "balance_loss_mlp": 1.02425313,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.0310914024365707,
      "language_loss": 0.72334743,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74609423,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.6031455993652344
    },
    {
      "auxiliary_loss_clip": 0.01201928,
      "auxiliary_loss_mlp": 0.01041505,
      "balance_loss_clip": 1.06329465,
      "balance_loss_mlp": 1.03007841,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.5476241457705995,
      "language_loss": 0.82207274,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.8445071,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.7037253379821777
    },
    {
      "auxiliary_loss_clip": 0.01226897,
      "auxiliary_loss_mlp": 0.01037225,
      "balance_loss_clip": 1.06738043,
      "balance_loss_mlp": 1.025841,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 2.2078538872498616,
      "language_loss": 0.8766768,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89931798,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.6897201538085938
    },
    {
      "auxiliary_loss_clip": 0.01231378,
      "auxiliary_loss_mlp": 0.01041818,
      "balance_loss_clip": 1.06708753,
      "balance_loss_mlp": 1.02968812,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.954812088725513,
      "language_loss": 0.81005758,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.83278954,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.6341910362243652
    },
    {
      "auxiliary_loss_clip": 0.01179022,
      "auxiliary_loss_mlp": 0.01044999,
      "balance_loss_clip": 1.05394197,
      "balance_loss_mlp": 1.03342938,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 1.8382370323017907,
      "language_loss": 0.8300705,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.85231078,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 3.6857733726501465
    },
    {
      "auxiliary_loss_clip": 0.0118542,
      "auxiliary_loss_mlp": 0.01037856,
      "balance_loss_clip": 1.05792224,
      "balance_loss_mlp": 1.02662611,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 1.9752719250963982,
      "language_loss": 0.87330973,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.8955425,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 3.6358978748321533
    },
    {
      "auxiliary_loss_clip": 0.01169104,
      "auxiliary_loss_mlp": 0.0104708,
      "balance_loss_clip": 1.05174232,
      "balance_loss_mlp": 1.03374624,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.3927660159083253,
      "language_loss": 0.97458315,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99674493,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.710602045059204
    },
    {
      "auxiliary_loss_clip": 0.01232214,
      "auxiliary_loss_mlp": 0.01036954,
      "balance_loss_clip": 1.06749725,
      "balance_loss_mlp": 1.02668393,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 4.190154509692528,
      "language_loss": 0.92462993,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94732165,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.567906141281128
    },
    {
      "auxiliary_loss_clip": 0.0120689,
      "auxiliary_loss_mlp": 0.01039199,
      "balance_loss_clip": 1.05552173,
      "balance_loss_mlp": 1.02703977,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 2.6788194606671034,
      "language_loss": 0.7413761,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.76383704,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.8946430683135986
    },
    {
      "auxiliary_loss_clip": 0.01234898,
      "auxiliary_loss_mlp": 0.01044538,
      "balance_loss_clip": 1.06823468,
      "balance_loss_mlp": 1.03164577,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.113857155812971,
      "language_loss": 0.91999179,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94278616,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.6486051082611084
    },
    {
      "auxiliary_loss_clip": 0.01097878,
      "auxiliary_loss_mlp": 0.01006879,
      "balance_loss_clip": 1.01887679,
      "balance_loss_mlp": 1.00380325,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.865658682051476,
      "language_loss": 0.62802011,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64906764,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.2087883949279785
    },
    {
      "auxiliary_loss_clip": 0.01186141,
      "auxiliary_loss_mlp": 0.01048032,
      "balance_loss_clip": 1.05920815,
      "balance_loss_mlp": 1.03592587,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 3.4483924354628943,
      "language_loss": 0.81454366,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83688533,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.717573404312134
    },
    {
      "auxiliary_loss_clip": 0.01236305,
      "auxiliary_loss_mlp": 0.01041605,
      "balance_loss_clip": 1.07166648,
      "balance_loss_mlp": 1.03134704,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.4265702953909845,
      "language_loss": 0.75788158,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.78066063,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.668071985244751
    },
    {
      "auxiliary_loss_clip": 0.01223565,
      "auxiliary_loss_mlp": 0.01043784,
      "balance_loss_clip": 1.05964255,
      "balance_loss_mlp": 1.03155303,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.6394949965594696,
      "language_loss": 0.80692565,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.82959914,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.720780372619629
    },
    {
      "auxiliary_loss_clip": 0.01243403,
      "auxiliary_loss_mlp": 0.00904607,
      "balance_loss_clip": 1.06874061,
      "balance_loss_mlp": 1.00099134,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.5698726704291577,
      "language_loss": 0.7857976,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80727768,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.6085574626922607
    },
    {
      "auxiliary_loss_clip": 0.01206084,
      "auxiliary_loss_mlp": 0.01044702,
      "balance_loss_clip": 1.05939817,
      "balance_loss_mlp": 1.03315079,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.977766954853013,
      "language_loss": 0.86318797,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88569582,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.8360023498535156
    },
    {
      "auxiliary_loss_clip": 0.01239944,
      "auxiliary_loss_mlp": 0.01042257,
      "balance_loss_clip": 1.06656313,
      "balance_loss_mlp": 1.0311228,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.588904240458221,
      "language_loss": 0.84930003,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.87212205,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.6047468185424805
    },
    {
      "auxiliary_loss_clip": 0.01242469,
      "auxiliary_loss_mlp": 0.01037243,
      "balance_loss_clip": 1.06893134,
      "balance_loss_mlp": 1.02530432,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 2.336230483973442,
      "language_loss": 0.79198456,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81478167,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.648738384246826
    },
    {
      "auxiliary_loss_clip": 0.01231357,
      "auxiliary_loss_mlp": 0.01039788,
      "balance_loss_clip": 1.06385517,
      "balance_loss_mlp": 1.02820086,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 3.121440099290512,
      "language_loss": 0.78245193,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80516338,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.6284563541412354
    },
    {
      "auxiliary_loss_clip": 0.01240719,
      "auxiliary_loss_mlp": 0.01042779,
      "balance_loss_clip": 1.06684613,
      "balance_loss_mlp": 1.03149581,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 2.4110207506664416,
      "language_loss": 0.7792359,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80207086,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.7233521938323975
    },
    {
      "auxiliary_loss_clip": 0.01222485,
      "auxiliary_loss_mlp": 0.00905047,
      "balance_loss_clip": 1.06192994,
      "balance_loss_mlp": 1.0011605,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 2.6264630277044887,
      "language_loss": 0.81875414,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.84002948,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.7146453857421875
    },
    {
      "auxiliary_loss_clip": 0.01206511,
      "auxiliary_loss_mlp": 0.01041079,
      "balance_loss_clip": 1.06288648,
      "balance_loss_mlp": 1.02987981,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 2.202159946617453,
      "language_loss": 0.75029802,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77277392,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.6622352600097656
    },
    {
      "auxiliary_loss_clip": 0.01236311,
      "auxiliary_loss_mlp": 0.01043437,
      "balance_loss_clip": 1.06539869,
      "balance_loss_mlp": 1.03150368,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.1610778045168617,
      "language_loss": 0.8501693,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87296677,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.6961171627044678
    },
    {
      "auxiliary_loss_clip": 0.01222535,
      "auxiliary_loss_mlp": 0.01044951,
      "balance_loss_clip": 1.05953288,
      "balance_loss_mlp": 1.03418648,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.3655242124056266,
      "language_loss": 0.55863035,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.58130515,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.6834497451782227
    },
    {
      "auxiliary_loss_clip": 0.01237537,
      "auxiliary_loss_mlp": 0.01037892,
      "balance_loss_clip": 1.06491745,
      "balance_loss_mlp": 1.02682328,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 2.3229198655285357,
      "language_loss": 0.87299728,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.8957516,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.6823716163635254
    },
    {
      "auxiliary_loss_clip": 0.01201316,
      "auxiliary_loss_mlp": 0.0103992,
      "balance_loss_clip": 1.05833077,
      "balance_loss_mlp": 1.02824962,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 2.0743390595572695,
      "language_loss": 0.82288206,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84529436,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.713218927383423
    },
    {
      "auxiliary_loss_clip": 0.01235675,
      "auxiliary_loss_mlp": 0.01043138,
      "balance_loss_clip": 1.06697011,
      "balance_loss_mlp": 1.03076994,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 2.3086671751882855,
      "language_loss": 0.91148275,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93427092,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.619509696960449
    },
    {
      "auxiliary_loss_clip": 0.01239654,
      "auxiliary_loss_mlp": 0.01038492,
      "balance_loss_clip": 1.06566477,
      "balance_loss_mlp": 1.0271914,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.7326327237330932,
      "language_loss": 0.83850348,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86128485,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.611638307571411
    },
    {
      "auxiliary_loss_clip": 0.01210759,
      "auxiliary_loss_mlp": 0.01042355,
      "balance_loss_clip": 1.05779469,
      "balance_loss_mlp": 1.0300107,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.71383299229439,
      "language_loss": 0.83775437,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86028552,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.771758794784546
    },
    {
      "auxiliary_loss_clip": 0.01219382,
      "auxiliary_loss_mlp": 0.01038283,
      "balance_loss_clip": 1.06584585,
      "balance_loss_mlp": 1.02733409,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 1.8898722586658818,
      "language_loss": 0.84299123,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86556786,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.6416356563568115
    },
    {
      "auxiliary_loss_clip": 0.01226818,
      "auxiliary_loss_mlp": 0.0104138,
      "balance_loss_clip": 1.06200576,
      "balance_loss_mlp": 1.03039503,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.8485432606019314,
      "language_loss": 0.69533026,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71801221,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.6979222297668457
    },
    {
      "auxiliary_loss_clip": 0.01101773,
      "auxiliary_loss_mlp": 0.01004224,
      "balance_loss_clip": 1.01958275,
      "balance_loss_mlp": 1.00110042,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.9231519853008134,
      "language_loss": 0.60914439,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.63020438,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 4.175241231918335
    },
    {
      "auxiliary_loss_clip": 0.01211936,
      "auxiliary_loss_mlp": 0.01036584,
      "balance_loss_clip": 1.06039703,
      "balance_loss_mlp": 1.02517533,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 2.4106255983671376,
      "language_loss": 0.74788582,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77037108,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 3.526810884475708
    },
    {
      "auxiliary_loss_clip": 0.01242888,
      "auxiliary_loss_mlp": 0.01044484,
      "balance_loss_clip": 1.06837928,
      "balance_loss_mlp": 1.03250957,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 3.6996295032705713,
      "language_loss": 0.85464746,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87752116,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 3.538149833679199
    },
    {
      "auxiliary_loss_clip": 0.01220012,
      "auxiliary_loss_mlp": 0.01047646,
      "balance_loss_clip": 1.0626682,
      "balance_loss_mlp": 1.03466439,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.6218829870679876,
      "language_loss": 0.74810004,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.77077663,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.7392921447753906
    },
    {
      "auxiliary_loss_clip": 0.01242314,
      "auxiliary_loss_mlp": 0.01044351,
      "balance_loss_clip": 1.06909525,
      "balance_loss_mlp": 1.03205442,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.220261033686588,
      "language_loss": 0.89691496,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91978157,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.5532758235931396
    },
    {
      "auxiliary_loss_clip": 0.01231281,
      "auxiliary_loss_mlp": 0.010392,
      "balance_loss_clip": 1.06620252,
      "balance_loss_mlp": 1.0264864,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.332823272323092,
      "language_loss": 0.81751299,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84021783,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.645130157470703
    },
    {
      "auxiliary_loss_clip": 0.0122885,
      "auxiliary_loss_mlp": 0.01038222,
      "balance_loss_clip": 1.06239688,
      "balance_loss_mlp": 1.02754068,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 1.8093579067393115,
      "language_loss": 0.81755424,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.84022492,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.614415407180786
    },
    {
      "auxiliary_loss_clip": 0.01239819,
      "auxiliary_loss_mlp": 0.01034833,
      "balance_loss_clip": 1.06652665,
      "balance_loss_mlp": 1.02272177,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.836264553265802,
      "language_loss": 0.81102085,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83376741,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.6660313606262207
    },
    {
      "auxiliary_loss_clip": 0.01110276,
      "auxiliary_loss_mlp": 0.01011653,
      "balance_loss_clip": 1.02597117,
      "balance_loss_mlp": 1.00911355,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.7000124678992372,
      "language_loss": 0.52635896,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54757822,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.121320962905884
    },
    {
      "auxiliary_loss_clip": 0.01206689,
      "auxiliary_loss_mlp": 0.01042978,
      "balance_loss_clip": 1.06202698,
      "balance_loss_mlp": 1.02927458,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.606176487839123,
      "language_loss": 0.8139385,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.8364352,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.7855827808380127
    },
    {
      "auxiliary_loss_clip": 0.01212449,
      "auxiliary_loss_mlp": 0.01036128,
      "balance_loss_clip": 1.06253123,
      "balance_loss_mlp": 1.02451134,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 5.418124302231683,
      "language_loss": 0.79416245,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.81664819,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.7544150352478027
    },
    {
      "auxiliary_loss_clip": 0.01204991,
      "auxiliary_loss_mlp": 0.01045339,
      "balance_loss_clip": 1.06068444,
      "balance_loss_mlp": 1.03347743,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.1822126881889568,
      "language_loss": 0.83011401,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85261738,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.7677993774414062
    },
    {
      "auxiliary_loss_clip": 0.01182617,
      "auxiliary_loss_mlp": 0.01044101,
      "balance_loss_clip": 1.05478764,
      "balance_loss_mlp": 1.03213882,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.8477995679820005,
      "language_loss": 0.8819325,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90419972,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.88065242767334
    },
    {
      "auxiliary_loss_clip": 0.0120568,
      "auxiliary_loss_mlp": 0.01038394,
      "balance_loss_clip": 1.05691445,
      "balance_loss_mlp": 1.02495885,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 7.6670690050125625,
      "language_loss": 0.7660414,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.78848219,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.7507848739624023
    },
    {
      "auxiliary_loss_clip": 0.0121499,
      "auxiliary_loss_mlp": 0.01044463,
      "balance_loss_clip": 1.06111789,
      "balance_loss_mlp": 1.0336324,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.81273876503136,
      "language_loss": 0.82687855,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.849473,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.692741632461548
    },
    {
      "auxiliary_loss_clip": 0.01227043,
      "auxiliary_loss_mlp": 0.01052132,
      "balance_loss_clip": 1.06146884,
      "balance_loss_mlp": 1.03989518,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 2.027473150552973,
      "language_loss": 0.83447719,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85726893,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.737910747528076
    },
    {
      "auxiliary_loss_clip": 0.01214298,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.05875278,
      "balance_loss_mlp": 1.02413797,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.219438351315837,
      "language_loss": 0.78590697,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80840945,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.676713228225708
    },
    {
      "auxiliary_loss_clip": 0.01216996,
      "auxiliary_loss_mlp": 0.01041582,
      "balance_loss_clip": 1.06235719,
      "balance_loss_mlp": 1.02974486,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 2.6921983066113437,
      "language_loss": 0.81097579,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.8335616,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.6697304248809814
    },
    {
      "auxiliary_loss_clip": 0.01234154,
      "auxiliary_loss_mlp": 0.01039165,
      "balance_loss_clip": 1.06374228,
      "balance_loss_mlp": 1.02657044,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 1.880880712699317,
      "language_loss": 0.74750173,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77023494,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.7861998081207275
    },
    {
      "auxiliary_loss_clip": 0.01183973,
      "auxiliary_loss_mlp": 0.01043705,
      "balance_loss_clip": 1.0528512,
      "balance_loss_mlp": 1.03208852,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 5.779518311156249,
      "language_loss": 0.86047375,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88275051,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.7604711055755615
    },
    {
      "auxiliary_loss_clip": 0.01183521,
      "auxiliary_loss_mlp": 0.01040855,
      "balance_loss_clip": 1.05368996,
      "balance_loss_mlp": 1.02916074,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 2.4335713704456863,
      "language_loss": 0.77019495,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79243875,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.7501473426818848
    },
    {
      "auxiliary_loss_clip": 0.01200255,
      "auxiliary_loss_mlp": 0.01040607,
      "balance_loss_clip": 1.06277657,
      "balance_loss_mlp": 1.0294075,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.1030291190299497,
      "language_loss": 0.8041572,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.82656586,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.7929790019989014
    },
    {
      "auxiliary_loss_clip": 0.01243693,
      "auxiliary_loss_mlp": 0.01049595,
      "balance_loss_clip": 1.06981015,
      "balance_loss_mlp": 1.03825831,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 2.2753668316665863,
      "language_loss": 0.79331148,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81624436,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.623328924179077
    },
    {
      "auxiliary_loss_clip": 0.0123056,
      "auxiliary_loss_mlp": 0.01039195,
      "balance_loss_clip": 1.06639004,
      "balance_loss_mlp": 1.02788758,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.93320239811219,
      "language_loss": 0.88215578,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90485334,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.636852264404297
    },
    {
      "auxiliary_loss_clip": 0.01223809,
      "auxiliary_loss_mlp": 0.00905469,
      "balance_loss_clip": 1.06361306,
      "balance_loss_mlp": 1.00167763,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 2.1324126848733984,
      "language_loss": 0.75279474,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.77408755,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.6410529613494873
    },
    {
      "auxiliary_loss_clip": 0.01230082,
      "auxiliary_loss_mlp": 0.01037139,
      "balance_loss_clip": 1.0624032,
      "balance_loss_mlp": 1.02530718,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 2.8092555552001666,
      "language_loss": 0.79390675,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.81657898,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.6520657539367676
    },
    {
      "auxiliary_loss_clip": 0.01225054,
      "auxiliary_loss_mlp": 0.01039422,
      "balance_loss_clip": 1.06249976,
      "balance_loss_mlp": 1.02627945,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.525950355195426,
      "language_loss": 0.91440284,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.9370476,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 4.5470287799835205
    },
    {
      "auxiliary_loss_clip": 0.01115206,
      "auxiliary_loss_mlp": 0.01007396,
      "balance_loss_clip": 1.0229485,
      "balance_loss_mlp": 1.00463057,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.783724437335986,
      "language_loss": 0.60909486,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63032091,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.0290801525115967
    },
    {
      "auxiliary_loss_clip": 0.01160237,
      "auxiliary_loss_mlp": 0.0104593,
      "balance_loss_clip": 1.05261898,
      "balance_loss_mlp": 1.03409815,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.0916117382250206,
      "language_loss": 0.73017418,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.75223589,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 3.8149144649505615
    },
    {
      "auxiliary_loss_clip": 0.0120738,
      "auxiliary_loss_mlp": 0.01043324,
      "balance_loss_clip": 1.06187773,
      "balance_loss_mlp": 1.03217208,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 2.309073558759019,
      "language_loss": 0.89054483,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91305184,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 3.691204309463501
    },
    {
      "auxiliary_loss_clip": 0.01206934,
      "auxiliary_loss_mlp": 0.01047655,
      "balance_loss_clip": 1.0601027,
      "balance_loss_mlp": 1.03605604,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.131121146861056,
      "language_loss": 0.71628356,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.73882937,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.7425103187561035
    },
    {
      "auxiliary_loss_clip": 0.01198093,
      "auxiliary_loss_mlp": 0.01041085,
      "balance_loss_clip": 1.05943716,
      "balance_loss_mlp": 1.03069592,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.064276482045777,
      "language_loss": 0.76912975,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79152149,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.721015691757202
    },
    {
      "auxiliary_loss_clip": 0.01201678,
      "auxiliary_loss_mlp": 0.01042057,
      "balance_loss_clip": 1.05566382,
      "balance_loss_mlp": 1.02972436,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 3.914365085521052,
      "language_loss": 0.76586795,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.78830534,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.767165184020996
    },
    {
      "auxiliary_loss_clip": 0.01209559,
      "auxiliary_loss_mlp": 0.01034963,
      "balance_loss_clip": 1.06283987,
      "balance_loss_mlp": 1.02388263,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 2.751882986508476,
      "language_loss": 0.71587884,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73832405,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.714104175567627
    },
    {
      "auxiliary_loss_clip": 0.01113782,
      "auxiliary_loss_mlp": 0.01002203,
      "balance_loss_clip": 1.02416086,
      "balance_loss_mlp": 0.99916327,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.90662510923579,
      "language_loss": 0.59087938,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61203927,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.3175694942474365
    },
    {
      "auxiliary_loss_clip": 0.01205022,
      "auxiliary_loss_mlp": 0.01038084,
      "balance_loss_clip": 1.05912137,
      "balance_loss_mlp": 1.02689612,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.6208471828884465,
      "language_loss": 0.81784254,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.84027356,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 2.9154446125030518
    },
    {
      "auxiliary_loss_clip": 0.01116969,
      "auxiliary_loss_mlp": 0.01001926,
      "balance_loss_clip": 1.02068686,
      "balance_loss_mlp": 0.9989931,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9575671651435397,
      "language_loss": 0.60754406,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62873304,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.1230897903442383
    },
    {
      "auxiliary_loss_clip": 0.01208516,
      "auxiliary_loss_mlp": 0.010398,
      "balance_loss_clip": 1.06349516,
      "balance_loss_mlp": 1.02849936,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 6.841443752627106,
      "language_loss": 0.77042413,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.7929073,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.7030537128448486
    },
    {
      "auxiliary_loss_clip": 0.01225199,
      "auxiliary_loss_mlp": 0.00905826,
      "balance_loss_clip": 1.06606448,
      "balance_loss_mlp": 1.00183249,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 12.300652764555712,
      "language_loss": 0.82362509,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84493536,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.7025210857391357
    },
    {
      "auxiliary_loss_clip": 0.01079838,
      "auxiliary_loss_mlp": 0.01009355,
      "balance_loss_clip": 1.02001834,
      "balance_loss_mlp": 1.00686359,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9392373856652823,
      "language_loss": 0.59461224,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61550421,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.117091178894043
    },
    {
      "auxiliary_loss_clip": 0.0121602,
      "auxiliary_loss_mlp": 0.01039031,
      "balance_loss_clip": 1.06352544,
      "balance_loss_mlp": 1.02688336,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 3.3903090557815796,
      "language_loss": 0.72456276,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74711323,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.7488744258880615
    },
    {
      "auxiliary_loss_clip": 0.0121907,
      "auxiliary_loss_mlp": 0.01044495,
      "balance_loss_clip": 1.06107235,
      "balance_loss_mlp": 1.03170395,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 3.1807706318439064,
      "language_loss": 0.68727964,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.70991528,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.7128372192382812
    },
    {
      "auxiliary_loss_clip": 0.01244052,
      "auxiliary_loss_mlp": 0.01037733,
      "balance_loss_clip": 1.06913662,
      "balance_loss_mlp": 1.02687323,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 2.983932545769554,
      "language_loss": 0.67631721,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.69913512,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.585280179977417
    },
    {
      "auxiliary_loss_clip": 0.012358,
      "auxiliary_loss_mlp": 0.01035105,
      "balance_loss_clip": 1.07112169,
      "balance_loss_mlp": 1.02355945,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.7925170936395831,
      "language_loss": 0.84200752,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86471653,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.6556613445281982
    },
    {
      "auxiliary_loss_clip": 0.01213429,
      "auxiliary_loss_mlp": 0.01046192,
      "balance_loss_clip": 1.05938327,
      "balance_loss_mlp": 1.0342952,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 2.088930346966219,
      "language_loss": 0.8749305,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.89752674,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.6608939170837402
    },
    {
      "auxiliary_loss_clip": 0.01233198,
      "auxiliary_loss_mlp": 0.01036493,
      "balance_loss_clip": 1.07092261,
      "balance_loss_mlp": 1.02560341,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 1.9552154630333092,
      "language_loss": 0.69648039,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71917731,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.641846179962158
    },
    {
      "auxiliary_loss_clip": 0.01189197,
      "auxiliary_loss_mlp": 0.01038584,
      "balance_loss_clip": 1.05888987,
      "balance_loss_mlp": 1.02660894,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 3.8010125180962,
      "language_loss": 0.92101622,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94329393,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.725266933441162
    },
    {
      "auxiliary_loss_clip": 0.0120781,
      "auxiliary_loss_mlp": 0.01039986,
      "balance_loss_clip": 1.05886424,
      "balance_loss_mlp": 1.02916217,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 3.0229284884601775,
      "language_loss": 0.75451398,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77699196,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.677246570587158
    },
    {
      "auxiliary_loss_clip": 0.01243562,
      "auxiliary_loss_mlp": 0.01040017,
      "balance_loss_clip": 1.06826425,
      "balance_loss_mlp": 1.02904379,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.6460436124198774,
      "language_loss": 0.78354967,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80638552,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.5822455883026123
    },
    {
      "auxiliary_loss_clip": 0.01248689,
      "auxiliary_loss_mlp": 0.01042108,
      "balance_loss_clip": 1.06902432,
      "balance_loss_mlp": 1.02936506,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.676552457096602,
      "language_loss": 0.84661329,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86952126,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.5774340629577637
    },
    {
      "auxiliary_loss_clip": 0.01213762,
      "auxiliary_loss_mlp": 0.00905479,
      "balance_loss_clip": 1.06363702,
      "balance_loss_mlp": 1.00175095,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.196465554135806,
      "language_loss": 0.91097605,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.93216842,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.6894209384918213
    },
    {
      "auxiliary_loss_clip": 0.01221612,
      "auxiliary_loss_mlp": 0.01039978,
      "balance_loss_clip": 1.06683564,
      "balance_loss_mlp": 1.0285759,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.7140735581307474,
      "language_loss": 0.90606809,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.928684,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.7489113807678223
    },
    {
      "auxiliary_loss_clip": 0.01110184,
      "auxiliary_loss_mlp": 0.01011642,
      "balance_loss_clip": 1.01921785,
      "balance_loss_mlp": 1.00872183,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.8283017848826186,
      "language_loss": 0.565727,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58694524,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 4.257274389266968
    },
    {
      "auxiliary_loss_clip": 0.01233906,
      "auxiliary_loss_mlp": 0.01045319,
      "balance_loss_clip": 1.06952882,
      "balance_loss_mlp": 1.03403556,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 2.0721327257312123,
      "language_loss": 0.78036547,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80315781,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 3.933176279067993
    },
    {
      "auxiliary_loss_clip": 0.01245066,
      "auxiliary_loss_mlp": 0.01043395,
      "balance_loss_clip": 1.07138395,
      "balance_loss_mlp": 1.03200483,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 4.218715072305209,
      "language_loss": 0.81806219,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.84094679,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 3.599919080734253
    },
    {
      "auxiliary_loss_clip": 0.01222211,
      "auxiliary_loss_mlp": 0.01036386,
      "balance_loss_clip": 1.06874728,
      "balance_loss_mlp": 1.02494192,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 7.225481994990703,
      "language_loss": 0.84347236,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86605829,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 3.612076759338379
    },
    {
      "auxiliary_loss_clip": 0.01248811,
      "auxiliary_loss_mlp": 0.01045514,
      "balance_loss_clip": 1.07241178,
      "balance_loss_mlp": 1.03270507,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 2.129376948999308,
      "language_loss": 0.73564315,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.75858641,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.641676187515259
    },
    {
      "auxiliary_loss_clip": 0.01091966,
      "auxiliary_loss_mlp": 0.0100245,
      "balance_loss_clip": 1.01827431,
      "balance_loss_mlp": 0.99961263,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8337826333546449,
      "language_loss": 0.54586077,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56680495,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.1800239086151123
    },
    {
      "auxiliary_loss_clip": 0.01239623,
      "auxiliary_loss_mlp": 0.01046867,
      "balance_loss_clip": 1.07100081,
      "balance_loss_mlp": 1.03455317,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.471630113861509,
      "language_loss": 0.69894582,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.7218107,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.7256197929382324
    },
    {
      "auxiliary_loss_clip": 0.01098635,
      "auxiliary_loss_mlp": 0.01002982,
      "balance_loss_clip": 1.01775956,
      "balance_loss_mlp": 1.00008476,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7593247295292854,
      "language_loss": 0.58076382,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60177994,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.157148599624634
    },
    {
      "auxiliary_loss_clip": 0.01235853,
      "auxiliary_loss_mlp": 0.00905708,
      "balance_loss_clip": 1.06698883,
      "balance_loss_mlp": 1.0015099,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.31282117237728,
      "language_loss": 0.73704243,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.75845802,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.775167226791382
    },
    {
      "auxiliary_loss_clip": 0.01178437,
      "auxiliary_loss_mlp": 0.01042051,
      "balance_loss_clip": 1.05547249,
      "balance_loss_mlp": 1.02950978,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 1.8883611390044828,
      "language_loss": 0.89957178,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.92177665,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.8085200786590576
    },
    {
      "auxiliary_loss_clip": 0.01200495,
      "auxiliary_loss_mlp": 0.01036811,
      "balance_loss_clip": 1.06294203,
      "balance_loss_mlp": 1.02520609,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 16.63108431860618,
      "language_loss": 0.8261385,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84851158,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.6985113620758057
    },
    {
      "auxiliary_loss_clip": 0.01231208,
      "auxiliary_loss_mlp": 0.01037412,
      "balance_loss_clip": 1.06593943,
      "balance_loss_mlp": 1.02606297,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 2.1213322780968786,
      "language_loss": 0.92588145,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94856763,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.6804914474487305
    },
    {
      "auxiliary_loss_clip": 0.01165493,
      "auxiliary_loss_mlp": 0.01042663,
      "balance_loss_clip": 1.0534699,
      "balance_loss_mlp": 1.03044987,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 2.4896255137255836,
      "language_loss": 0.78720129,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80928284,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.9185218811035156
    },
    {
      "auxiliary_loss_clip": 0.0124824,
      "auxiliary_loss_mlp": 0.01043489,
      "balance_loss_clip": 1.07128716,
      "balance_loss_mlp": 1.03172863,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.8635861488520766,
      "language_loss": 0.74176842,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76468569,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.6011199951171875
    },
    {
      "auxiliary_loss_clip": 0.0121167,
      "auxiliary_loss_mlp": 0.01051138,
      "balance_loss_clip": 1.06756926,
      "balance_loss_mlp": 1.03810275,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.9222684902902205,
      "language_loss": 0.8529048,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87553287,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.682697057723999
    },
    {
      "auxiliary_loss_clip": 0.01225563,
      "auxiliary_loss_mlp": 0.00905972,
      "balance_loss_clip": 1.06634712,
      "balance_loss_mlp": 1.00166535,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.3424325905137033,
      "language_loss": 0.91395169,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.93526709,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.6720330715179443
    },
    {
      "auxiliary_loss_clip": 0.01249304,
      "auxiliary_loss_mlp": 0.01045302,
      "balance_loss_clip": 1.07113051,
      "balance_loss_mlp": 1.03310072,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.620140931610471,
      "language_loss": 0.66602975,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68897581,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.5571091175079346
    },
    {
      "auxiliary_loss_clip": 0.01221103,
      "auxiliary_loss_mlp": 0.01035962,
      "balance_loss_clip": 1.06788611,
      "balance_loss_mlp": 1.02528048,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 4.228343171483672,
      "language_loss": 0.84914422,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87171489,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.6953766345977783
    },
    {
      "auxiliary_loss_clip": 0.01235332,
      "auxiliary_loss_mlp": 0.01051436,
      "balance_loss_clip": 1.06714475,
      "balance_loss_mlp": 1.03961682,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 2.068751069962532,
      "language_loss": 0.74597859,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76884615,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.863799810409546
    },
    {
      "auxiliary_loss_clip": 0.01237385,
      "auxiliary_loss_mlp": 0.01043157,
      "balance_loss_clip": 1.06683826,
      "balance_loss_mlp": 1.03080654,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 2.5765804960714798,
      "language_loss": 0.79207873,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81488413,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.7157657146453857
    },
    {
      "auxiliary_loss_clip": 0.0121113,
      "auxiliary_loss_mlp": 0.00905694,
      "balance_loss_clip": 1.06162894,
      "balance_loss_mlp": 1.00165844,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 2.863724432685016,
      "language_loss": 0.85765839,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.87882662,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.634962320327759
    },
    {
      "auxiliary_loss_clip": 0.01211352,
      "auxiliary_loss_mlp": 0.01051906,
      "balance_loss_clip": 1.06073236,
      "balance_loss_mlp": 1.03891206,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 1.7428541174284873,
      "language_loss": 0.85718399,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.87981665,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.7439701557159424
    },
    {
      "auxiliary_loss_clip": 0.01219672,
      "auxiliary_loss_mlp": 0.01042407,
      "balance_loss_clip": 1.06630158,
      "balance_loss_mlp": 1.03040242,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.7647897158318397,
      "language_loss": 0.74051523,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76313603,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.6284098625183105
    },
    {
      "auxiliary_loss_clip": 0.01202062,
      "auxiliary_loss_mlp": 0.01037231,
      "balance_loss_clip": 1.06341529,
      "balance_loss_mlp": 1.02595937,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.2344476489308063,
      "language_loss": 0.87308741,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.89548028,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.664760112762451
    },
    {
      "auxiliary_loss_clip": 0.012477,
      "auxiliary_loss_mlp": 0.01039931,
      "balance_loss_clip": 1.07234848,
      "balance_loss_mlp": 1.02850461,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 2.175798292554038,
      "language_loss": 0.69970584,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.7225821,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.5970873832702637
    },
    {
      "auxiliary_loss_clip": 0.01196429,
      "auxiliary_loss_mlp": 0.01049588,
      "balance_loss_clip": 1.06104255,
      "balance_loss_mlp": 1.03723752,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.8160750832769788,
      "language_loss": 0.84413993,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.8666001,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.701997995376587
    },
    {
      "auxiliary_loss_clip": 0.01222847,
      "auxiliary_loss_mlp": 0.01037992,
      "balance_loss_clip": 1.06792593,
      "balance_loss_mlp": 1.02562976,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 1.6919998304516917,
      "language_loss": 0.8431797,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.8657881,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.694136619567871
    },
    {
      "auxiliary_loss_clip": 0.01212823,
      "auxiliary_loss_mlp": 0.01046368,
      "balance_loss_clip": 1.06869745,
      "balance_loss_mlp": 1.03407168,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.8028782384289155,
      "language_loss": 0.76374459,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78633654,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 3.59822154045105
    },
    {
      "auxiliary_loss_clip": 0.01221056,
      "auxiliary_loss_mlp": 0.01039651,
      "balance_loss_clip": 1.07041883,
      "balance_loss_mlp": 1.02828431,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.7364250906615921,
      "language_loss": 0.74822009,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77082717,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.663655996322632
    },
    {
      "auxiliary_loss_clip": 0.01232768,
      "auxiliary_loss_mlp": 0.01043151,
      "balance_loss_clip": 1.07021797,
      "balance_loss_mlp": 1.03202915,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.8298133643603136,
      "language_loss": 0.90795165,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.93071091,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 3.5578606128692627
    },
    {
      "auxiliary_loss_clip": 0.0120241,
      "auxiliary_loss_mlp": 0.0104209,
      "balance_loss_clip": 1.06383443,
      "balance_loss_mlp": 1.03085458,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.562741308856548,
      "language_loss": 0.79414046,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81658548,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 3.587360143661499
    },
    {
      "auxiliary_loss_clip": 0.0119588,
      "auxiliary_loss_mlp": 0.01048182,
      "balance_loss_clip": 1.05985928,
      "balance_loss_mlp": 1.03621376,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.165822683712787,
      "language_loss": 0.84111309,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.8635537,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.7730493545532227
    },
    {
      "auxiliary_loss_clip": 0.01243423,
      "auxiliary_loss_mlp": 0.01045867,
      "balance_loss_clip": 1.07085264,
      "balance_loss_mlp": 1.03419077,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 1.7815439434748541,
      "language_loss": 0.8237955,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84668839,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.5538134574890137
    },
    {
      "auxiliary_loss_clip": 0.0124016,
      "auxiliary_loss_mlp": 0.01045879,
      "balance_loss_clip": 1.06787062,
      "balance_loss_mlp": 1.03411317,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.7253614601622391,
      "language_loss": 0.83245242,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85531282,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.5985267162323
    },
    {
      "auxiliary_loss_clip": 0.01223928,
      "auxiliary_loss_mlp": 0.01047125,
      "balance_loss_clip": 1.06971812,
      "balance_loss_mlp": 1.03499508,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.013441513988337,
      "language_loss": 0.87110609,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89381665,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.6547131538391113
    },
    {
      "auxiliary_loss_clip": 0.01237349,
      "auxiliary_loss_mlp": 0.01054387,
      "balance_loss_clip": 1.06855166,
      "balance_loss_mlp": 1.04167914,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 2.1477504242449674,
      "language_loss": 0.70205325,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72497058,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.683002471923828
    },
    {
      "auxiliary_loss_clip": 0.01230512,
      "auxiliary_loss_mlp": 0.00905749,
      "balance_loss_clip": 1.06518114,
      "balance_loss_mlp": 1.00196528,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 2.0263565429056523,
      "language_loss": 0.859626,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.8809886,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.6583070755004883
    },
    {
      "auxiliary_loss_clip": 0.01226523,
      "auxiliary_loss_mlp": 0.01047011,
      "balance_loss_clip": 1.06692815,
      "balance_loss_mlp": 1.03460765,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 1.9578078132789607,
      "language_loss": 0.77396667,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79670203,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.795419454574585
    },
    {
      "auxiliary_loss_clip": 0.01232964,
      "auxiliary_loss_mlp": 0.01043802,
      "balance_loss_clip": 1.06835103,
      "balance_loss_mlp": 1.03135657,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 74.0948201076389,
      "language_loss": 0.70664334,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72941089,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.7610671520233154
    },
    {
      "auxiliary_loss_clip": 0.01189122,
      "auxiliary_loss_mlp": 0.01040602,
      "balance_loss_clip": 1.05793881,
      "balance_loss_mlp": 1.02925956,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.0304793147108255,
      "language_loss": 0.79077291,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81307018,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.658236026763916
    },
    {
      "auxiliary_loss_clip": 0.01217367,
      "auxiliary_loss_mlp": 0.01045604,
      "balance_loss_clip": 1.06163073,
      "balance_loss_mlp": 1.03302133,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 3.177659940811365,
      "language_loss": 0.71316433,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.73579407,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.666565418243408
    },
    {
      "auxiliary_loss_clip": 0.0119794,
      "auxiliary_loss_mlp": 0.01043201,
      "balance_loss_clip": 1.06262934,
      "balance_loss_mlp": 1.03166199,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.6593817149868508,
      "language_loss": 0.80997145,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.8323828,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.8118841648101807
    },
    {
      "auxiliary_loss_clip": 0.0123618,
      "auxiliary_loss_mlp": 0.01045285,
      "balance_loss_clip": 1.0677557,
      "balance_loss_mlp": 1.03313744,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.8514263317224222,
      "language_loss": 0.83001274,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.85282737,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.6396143436431885
    },
    {
      "auxiliary_loss_clip": 0.0108825,
      "auxiliary_loss_mlp": 0.01015437,
      "balance_loss_clip": 1.01970267,
      "balance_loss_mlp": 1.01228988,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7904212043167789,
      "language_loss": 0.64722669,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66826355,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.354785680770874
    },
    {
      "auxiliary_loss_clip": 0.01212038,
      "auxiliary_loss_mlp": 0.01034705,
      "balance_loss_clip": 1.06746137,
      "balance_loss_mlp": 1.02392817,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.814694647407419,
      "language_loss": 0.79664195,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.81910938,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.70086932182312
    },
    {
      "auxiliary_loss_clip": 0.01213401,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.06579936,
      "balance_loss_mlp": 1.02497125,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 2.781519913094165,
      "language_loss": 0.85327631,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87578011,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.664924383163452
    },
    {
      "auxiliary_loss_clip": 0.01230958,
      "auxiliary_loss_mlp": 0.01050108,
      "balance_loss_clip": 1.06486559,
      "balance_loss_mlp": 1.0383122,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 2.076427013956876,
      "language_loss": 0.84939677,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.8722074,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.677774429321289
    },
    {
      "auxiliary_loss_clip": 0.01204147,
      "auxiliary_loss_mlp": 0.01038484,
      "balance_loss_clip": 1.06064355,
      "balance_loss_mlp": 1.02751637,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 4.322246169357338,
      "language_loss": 0.8323195,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85474575,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.817049741744995
    },
    {
      "auxiliary_loss_clip": 0.01244083,
      "auxiliary_loss_mlp": 0.01041023,
      "balance_loss_clip": 1.07176542,
      "balance_loss_mlp": 1.02837491,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.7749898545856089,
      "language_loss": 0.78981137,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81266242,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.642831325531006
    },
    {
      "auxiliary_loss_clip": 0.01223468,
      "auxiliary_loss_mlp": 0.00906778,
      "balance_loss_clip": 1.06213021,
      "balance_loss_mlp": 1.00211227,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.295439387403753,
      "language_loss": 0.88420272,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.90550518,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.6178929805755615
    },
    {
      "auxiliary_loss_clip": 0.0123724,
      "auxiliary_loss_mlp": 0.00906357,
      "balance_loss_clip": 1.06719863,
      "balance_loss_mlp": 1.00188816,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.767544170804333,
      "language_loss": 0.80288684,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82432282,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.710527181625366
    },
    {
      "auxiliary_loss_clip": 0.01218496,
      "auxiliary_loss_mlp": 0.0104331,
      "balance_loss_clip": 1.05913758,
      "balance_loss_mlp": 1.03067982,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.3207094557887524,
      "language_loss": 0.82596111,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.84857917,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.682162284851074
    },
    {
      "auxiliary_loss_clip": 0.01221337,
      "auxiliary_loss_mlp": 0.01029906,
      "balance_loss_clip": 1.07002616,
      "balance_loss_mlp": 1.01876605,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 2.2151597508842027,
      "language_loss": 0.83199787,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.85451031,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.6208579540252686
    },
    {
      "auxiliary_loss_clip": 0.01233718,
      "auxiliary_loss_mlp": 0.01050303,
      "balance_loss_clip": 1.06783831,
      "balance_loss_mlp": 1.03729737,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 3.234363051068786,
      "language_loss": 0.69049835,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71333849,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 3.9284703731536865
    },
    {
      "auxiliary_loss_clip": 0.01203192,
      "auxiliary_loss_mlp": 0.01038202,
      "balance_loss_clip": 1.06162429,
      "balance_loss_mlp": 1.02732992,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.236151879747322,
      "language_loss": 0.86900306,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89141697,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 3.6249890327453613
    },
    {
      "auxiliary_loss_clip": 0.01205165,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.06083703,
      "balance_loss_mlp": 1.02217269,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 2.7449466858398206,
      "language_loss": 0.75638932,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.77877927,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.7132744789123535
    },
    {
      "auxiliary_loss_clip": 0.01236192,
      "auxiliary_loss_mlp": 0.01051481,
      "balance_loss_clip": 1.06985331,
      "balance_loss_mlp": 1.03911281,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.293954835524856,
      "language_loss": 0.93790525,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.96078199,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 3.524379014968872
    },
    {
      "auxiliary_loss_clip": 0.01217169,
      "auxiliary_loss_mlp": 0.01045975,
      "balance_loss_clip": 1.0682162,
      "balance_loss_mlp": 1.0340476,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 1.8764480543509952,
      "language_loss": 0.82376724,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84639871,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 3.7049005031585693
    },
    {
      "auxiliary_loss_clip": 0.01212897,
      "auxiliary_loss_mlp": 0.0104264,
      "balance_loss_clip": 1.06717706,
      "balance_loss_mlp": 1.03048623,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 3.0999371056085145,
      "language_loss": 0.7591278,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.78168321,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.681532382965088
    },
    {
      "auxiliary_loss_clip": 0.0123123,
      "auxiliary_loss_mlp": 0.01042409,
      "balance_loss_clip": 1.06977558,
      "balance_loss_mlp": 1.03017783,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 2.67645317068984,
      "language_loss": 0.78853536,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.81127179,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.7198805809020996
    },
    {
      "auxiliary_loss_clip": 0.01101938,
      "auxiliary_loss_mlp": 0.01003166,
      "balance_loss_clip": 1.02698183,
      "balance_loss_mlp": 1.00006628,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7704245386512756,
      "language_loss": 0.55653608,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57758713,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.196312665939331
    },
    {
      "auxiliary_loss_clip": 0.01234431,
      "auxiliary_loss_mlp": 0.01048285,
      "balance_loss_clip": 1.06495547,
      "balance_loss_mlp": 1.03568482,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 4.606367532664015,
      "language_loss": 0.71522164,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73804885,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.655010938644409
    },
    {
      "auxiliary_loss_clip": 0.01226351,
      "auxiliary_loss_mlp": 0.01041437,
      "balance_loss_clip": 1.06195712,
      "balance_loss_mlp": 1.02791226,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.8903089570451694,
      "language_loss": 0.74396551,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.7666434,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.634387254714966
    },
    {
      "auxiliary_loss_clip": 0.01230801,
      "auxiliary_loss_mlp": 0.01054483,
      "balance_loss_clip": 1.06798649,
      "balance_loss_mlp": 1.0415132,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 3.0288516554873346,
      "language_loss": 0.72617435,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74902725,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.5705935955047607
    },
    {
      "auxiliary_loss_clip": 0.01229178,
      "auxiliary_loss_mlp": 0.01041644,
      "balance_loss_clip": 1.06916773,
      "balance_loss_mlp": 1.03048015,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.9988789710085377,
      "language_loss": 0.9651345,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98784274,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.6796891689300537
    },
    {
      "auxiliary_loss_clip": 0.01215921,
      "auxiliary_loss_mlp": 0.0104073,
      "balance_loss_clip": 1.06139147,
      "balance_loss_mlp": 1.02745581,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 2.517912655493947,
      "language_loss": 0.85720706,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.87977356,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.668612241744995
    },
    {
      "auxiliary_loss_clip": 0.01240196,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.06845737,
      "balance_loss_mlp": 1.02487731,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.4297117364543603,
      "language_loss": 0.81732088,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.84008336,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.6136581897735596
    },
    {
      "auxiliary_loss_clip": 0.01199567,
      "auxiliary_loss_mlp": 0.0103229,
      "balance_loss_clip": 1.05962372,
      "balance_loss_mlp": 1.02119207,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 2.045252817430676,
      "language_loss": 0.80094337,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82326186,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.666815757751465
    },
    {
      "auxiliary_loss_clip": 0.01228493,
      "auxiliary_loss_mlp": 0.01049388,
      "balance_loss_clip": 1.0651406,
      "balance_loss_mlp": 1.03669846,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 2.7861029206850905,
      "language_loss": 0.64818203,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.6709609,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.738725423812866
    },
    {
      "auxiliary_loss_clip": 0.01238607,
      "auxiliary_loss_mlp": 0.01041374,
      "balance_loss_clip": 1.06928933,
      "balance_loss_mlp": 1.03029394,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 1.9249790537109528,
      "language_loss": 0.85414708,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87694693,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.5793440341949463
    },
    {
      "auxiliary_loss_clip": 0.01178806,
      "auxiliary_loss_mlp": 0.01042755,
      "balance_loss_clip": 1.05716932,
      "balance_loss_mlp": 1.03120971,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 4.775638811835069,
      "language_loss": 0.8103317,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83254737,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.8683009147644043
    },
    {
      "auxiliary_loss_clip": 0.01225815,
      "auxiliary_loss_mlp": 0.01041944,
      "balance_loss_clip": 1.06576777,
      "balance_loss_mlp": 1.02934337,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.143728926256755,
      "language_loss": 0.823192,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84586954,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.6590752601623535
    },
    {
      "auxiliary_loss_clip": 0.01184158,
      "auxiliary_loss_mlp": 0.01045316,
      "balance_loss_clip": 1.05529523,
      "balance_loss_mlp": 1.03366351,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.8390765353991714,
      "language_loss": 0.65181786,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.67411256,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.7683160305023193
    },
    {
      "auxiliary_loss_clip": 0.01215934,
      "auxiliary_loss_mlp": 0.01045419,
      "balance_loss_clip": 1.06148219,
      "balance_loss_mlp": 1.03333747,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.3673567922629366,
      "language_loss": 0.88994491,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91255844,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.6845436096191406
    },
    {
      "auxiliary_loss_clip": 0.01224648,
      "auxiliary_loss_mlp": 0.01043875,
      "balance_loss_clip": 1.06461549,
      "balance_loss_mlp": 1.03238893,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.7903737485907365,
      "language_loss": 0.82106382,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84374905,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.629899740219116
    },
    {
      "auxiliary_loss_clip": 0.01216892,
      "auxiliary_loss_mlp": 0.01038249,
      "balance_loss_clip": 1.06548238,
      "balance_loss_mlp": 1.02636981,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 12.701549402754228,
      "language_loss": 0.80358529,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82613665,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.6407241821289062
    },
    {
      "auxiliary_loss_clip": 0.01172574,
      "auxiliary_loss_mlp": 0.01038129,
      "balance_loss_clip": 1.05470896,
      "balance_loss_mlp": 1.02658939,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 2.2175709251959623,
      "language_loss": 0.69450545,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71661246,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.8630969524383545
    },
    {
      "auxiliary_loss_clip": 0.01225764,
      "auxiliary_loss_mlp": 0.0104563,
      "balance_loss_clip": 1.06438625,
      "balance_loss_mlp": 1.0338763,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.7141977163843232,
      "language_loss": 0.78434139,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80705535,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.7292022705078125
    },
    {
      "auxiliary_loss_clip": 0.01217175,
      "auxiliary_loss_mlp": 0.00905493,
      "balance_loss_clip": 1.06631815,
      "balance_loss_mlp": 1.00182605,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 3.03000512591291,
      "language_loss": 0.77794039,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.79916704,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.649030923843384
    },
    {
      "auxiliary_loss_clip": 0.01197464,
      "auxiliary_loss_mlp": 0.01047416,
      "balance_loss_clip": 1.05996609,
      "balance_loss_mlp": 1.03434479,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.671854461371811,
      "language_loss": 0.78312629,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80557501,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.6944780349731445
    },
    {
      "auxiliary_loss_clip": 0.01245834,
      "auxiliary_loss_mlp": 0.00906668,
      "balance_loss_clip": 1.07120383,
      "balance_loss_mlp": 1.00189638,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.7046839005109486,
      "language_loss": 0.71321547,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.7347405,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 4.472017526626587
    },
    {
      "auxiliary_loss_clip": 0.01243949,
      "auxiliary_loss_mlp": 0.0104502,
      "balance_loss_clip": 1.06996024,
      "balance_loss_mlp": 1.0334928,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 11.486980226675518,
      "language_loss": 0.75576568,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77865541,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.715036630630493
    },
    {
      "auxiliary_loss_clip": 0.012156,
      "auxiliary_loss_mlp": 0.01048405,
      "balance_loss_clip": 1.0589267,
      "balance_loss_mlp": 1.03559613,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 3.1647761514257433,
      "language_loss": 0.89456159,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91720158,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 3.5508384704589844
    },
    {
      "auxiliary_loss_clip": 0.01217447,
      "auxiliary_loss_mlp": 0.01050694,
      "balance_loss_clip": 1.06547332,
      "balance_loss_mlp": 1.03834963,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 1.8642193107664078,
      "language_loss": 0.86177444,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.8844558,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 3.582043170928955
    },
    {
      "auxiliary_loss_clip": 0.01242024,
      "auxiliary_loss_mlp": 0.01045051,
      "balance_loss_clip": 1.07013524,
      "balance_loss_mlp": 1.03301072,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.3922281971080563,
      "language_loss": 0.73738605,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76025677,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.543853521347046
    },
    {
      "auxiliary_loss_clip": 0.01242794,
      "auxiliary_loss_mlp": 0.00905914,
      "balance_loss_clip": 1.06827974,
      "balance_loss_mlp": 1.0018158,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.9392485194183866,
      "language_loss": 0.85269839,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.8741855,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.581806182861328
    },
    {
      "auxiliary_loss_clip": 0.01230374,
      "auxiliary_loss_mlp": 0.01037504,
      "balance_loss_clip": 1.07086086,
      "balance_loss_mlp": 1.02597046,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 1.9578331355514502,
      "language_loss": 0.77964711,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.80232584,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.6748576164245605
    },
    {
      "auxiliary_loss_clip": 0.01227396,
      "auxiliary_loss_mlp": 0.01043196,
      "balance_loss_clip": 1.06589293,
      "balance_loss_mlp": 1.03108454,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.3377757316593666,
      "language_loss": 0.84420109,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86690706,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.7011497020721436
    },
    {
      "auxiliary_loss_clip": 0.01226073,
      "auxiliary_loss_mlp": 0.0104607,
      "balance_loss_clip": 1.0676713,
      "balance_loss_mlp": 1.03537107,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.0628398987398113,
      "language_loss": 0.74937868,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77210015,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.5467464923858643
    },
    {
      "auxiliary_loss_clip": 0.01203949,
      "auxiliary_loss_mlp": 0.01044195,
      "balance_loss_clip": 1.0611732,
      "balance_loss_mlp": 1.03322744,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 3.3866863122960438,
      "language_loss": 0.88668978,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.90917122,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.6959924697875977
    },
    {
      "auxiliary_loss_clip": 0.01233326,
      "auxiliary_loss_mlp": 0.01043919,
      "balance_loss_clip": 1.07042551,
      "balance_loss_mlp": 1.03137207,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.9235746820174076,
      "language_loss": 0.83788973,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.86066222,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.598501443862915
    },
    {
      "auxiliary_loss_clip": 0.01225006,
      "auxiliary_loss_mlp": 0.01038823,
      "balance_loss_clip": 1.06433904,
      "balance_loss_mlp": 1.02742708,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.6140421767424817,
      "language_loss": 0.74755108,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.7701894,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.823869228363037
    },
    {
      "auxiliary_loss_clip": 0.01239812,
      "auxiliary_loss_mlp": 0.01036844,
      "balance_loss_clip": 1.06849003,
      "balance_loss_mlp": 1.02547133,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 3.6682433790849265,
      "language_loss": 0.78573614,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80850267,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.639935255050659
    },
    {
      "auxiliary_loss_clip": 0.01245438,
      "auxiliary_loss_mlp": 0.01039572,
      "balance_loss_clip": 1.07006931,
      "balance_loss_mlp": 1.02691185,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.384006829602795,
      "language_loss": 0.70197022,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72482038,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.5622448921203613
    },
    {
      "auxiliary_loss_clip": 0.01214688,
      "auxiliary_loss_mlp": 0.01046454,
      "balance_loss_clip": 1.06800604,
      "balance_loss_mlp": 1.03463411,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.2899516322426385,
      "language_loss": 0.97078145,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.99339288,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.6896777153015137
    },
    {
      "auxiliary_loss_clip": 0.01212794,
      "auxiliary_loss_mlp": 0.01046742,
      "balance_loss_clip": 1.06130958,
      "balance_loss_mlp": 1.03638256,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 1.972833262475155,
      "language_loss": 0.73822927,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.76082462,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.6966443061828613
    },
    {
      "auxiliary_loss_clip": 0.01232255,
      "auxiliary_loss_mlp": 0.01042898,
      "balance_loss_clip": 1.07211721,
      "balance_loss_mlp": 1.03209162,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.3899304999556983,
      "language_loss": 0.77699977,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.79975128,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.6319401264190674
    },
    {
      "auxiliary_loss_clip": 0.01221586,
      "auxiliary_loss_mlp": 0.0104547,
      "balance_loss_clip": 1.06349993,
      "balance_loss_mlp": 1.03402567,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 2.1129372463201057,
      "language_loss": 0.7366513,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.75932181,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.6945319175720215
    },
    {
      "auxiliary_loss_clip": 0.01242222,
      "auxiliary_loss_mlp": 0.01041254,
      "balance_loss_clip": 1.06953394,
      "balance_loss_mlp": 1.02920175,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.475561464595169,
      "language_loss": 0.84256929,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86540407,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.6744632720947266
    },
    {
      "auxiliary_loss_clip": 0.01209419,
      "auxiliary_loss_mlp": 0.01038265,
      "balance_loss_clip": 1.05961668,
      "balance_loss_mlp": 1.0266602,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 4.851024700950319,
      "language_loss": 0.64194721,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66442406,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.7103378772735596
    },
    {
      "auxiliary_loss_clip": 0.01216375,
      "auxiliary_loss_mlp": 0.01042625,
      "balance_loss_clip": 1.06680632,
      "balance_loss_mlp": 1.0314666,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.557345935143308,
      "language_loss": 0.79686677,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.81945676,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.643540859222412
    },
    {
      "auxiliary_loss_clip": 0.01234233,
      "auxiliary_loss_mlp": 0.00905246,
      "balance_loss_clip": 1.07037807,
      "balance_loss_mlp": 1.00140262,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.4955596095163575,
      "language_loss": 0.8292402,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.85063505,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.625922679901123
    },
    {
      "auxiliary_loss_clip": 0.01236906,
      "auxiliary_loss_mlp": 0.01037314,
      "balance_loss_clip": 1.06859064,
      "balance_loss_mlp": 1.02642417,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 2.793448517807113,
      "language_loss": 0.70029116,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72303337,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.639695882797241
    },
    {
      "auxiliary_loss_clip": 0.01220689,
      "auxiliary_loss_mlp": 0.010449,
      "balance_loss_clip": 1.06913912,
      "balance_loss_mlp": 1.03442156,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.364095371191123,
      "language_loss": 0.89937258,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.92202854,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.5669620037078857
    },
    {
      "auxiliary_loss_clip": 0.01229127,
      "auxiliary_loss_mlp": 0.01044925,
      "balance_loss_clip": 1.06943357,
      "balance_loss_mlp": 1.03345728,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 2.7463651366879414,
      "language_loss": 0.83962929,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.86236989,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.634547710418701
    },
    {
      "auxiliary_loss_clip": 0.01218156,
      "auxiliary_loss_mlp": 0.01038156,
      "balance_loss_clip": 1.0661993,
      "balance_loss_mlp": 1.02706993,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.031472539997053,
      "language_loss": 0.72537482,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74793804,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.6877641677856445
    },
    {
      "auxiliary_loss_clip": 0.01190232,
      "auxiliary_loss_mlp": 0.01036681,
      "balance_loss_clip": 1.05588078,
      "balance_loss_mlp": 1.02569628,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.3982663543070766,
      "language_loss": 0.70123011,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72349924,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 2.8156814575195312
    },
    {
      "auxiliary_loss_clip": 0.01180748,
      "auxiliary_loss_mlp": 0.01033405,
      "balance_loss_clip": 1.05896378,
      "balance_loss_mlp": 1.02117467,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 2.5236499430411854,
      "language_loss": 0.89777678,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.9199183,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 3.7203381061553955
    },
    {
      "auxiliary_loss_clip": 0.0122264,
      "auxiliary_loss_mlp": 0.01041796,
      "balance_loss_clip": 1.06834912,
      "balance_loss_mlp": 1.03041124,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.256432672387913,
      "language_loss": 0.82988894,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.85253334,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.63681697845459
    },
    {
      "auxiliary_loss_clip": 0.01209325,
      "auxiliary_loss_mlp": 0.01040846,
      "balance_loss_clip": 1.05957723,
      "balance_loss_mlp": 1.02863884,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 2.5803229073262917,
      "language_loss": 0.78082931,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80333102,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 3.7641520500183105
    },
    {
      "auxiliary_loss_clip": 0.01228091,
      "auxiliary_loss_mlp": 0.01039684,
      "balance_loss_clip": 1.06560361,
      "balance_loss_mlp": 1.02967024,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.5353544269989419,
      "language_loss": 0.83640832,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85908604,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 3.6126091480255127
    },
    {
      "auxiliary_loss_clip": 0.01223178,
      "auxiliary_loss_mlp": 0.01034198,
      "balance_loss_clip": 1.06862485,
      "balance_loss_mlp": 1.02210414,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.3579937371286612,
      "language_loss": 0.7629813,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78555501,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.647322654724121
    },
    {
      "auxiliary_loss_clip": 0.01227368,
      "auxiliary_loss_mlp": 0.01044896,
      "balance_loss_clip": 1.0683279,
      "balance_loss_mlp": 1.03420329,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.0756446426019086,
      "language_loss": 0.82326066,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84598327,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.671506881713867
    },
    {
      "auxiliary_loss_clip": 0.01229565,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.06849694,
      "balance_loss_mlp": 1.02488554,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 3.792555643231012,
      "language_loss": 0.82397401,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84662914,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.6527836322784424
    },
    {
      "auxiliary_loss_clip": 0.01193138,
      "auxiliary_loss_mlp": 0.01043202,
      "balance_loss_clip": 1.05753958,
      "balance_loss_mlp": 1.03227067,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.47049731272787,
      "language_loss": 0.88093388,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90329731,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.737354040145874
    },
    {
      "auxiliary_loss_clip": 0.01198349,
      "auxiliary_loss_mlp": 0.01044735,
      "balance_loss_clip": 1.06038857,
      "balance_loss_mlp": 1.03411329,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.8672287171584996,
      "language_loss": 0.86983508,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.89226592,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.6629579067230225
    },
    {
      "auxiliary_loss_clip": 0.01217675,
      "auxiliary_loss_mlp": 0.01036978,
      "balance_loss_clip": 1.06409407,
      "balance_loss_mlp": 1.02586734,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.761043242739168,
      "language_loss": 0.74364138,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76618791,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.7315711975097656
    },
    {
      "auxiliary_loss_clip": 0.01236707,
      "auxiliary_loss_mlp": 0.01045151,
      "balance_loss_clip": 1.06999552,
      "balance_loss_mlp": 1.03466082,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 2.0947878270397013,
      "language_loss": 0.82657242,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.84939104,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.5853421688079834
    },
    {
      "auxiliary_loss_clip": 0.0119669,
      "auxiliary_loss_mlp": 0.01044444,
      "balance_loss_clip": 1.05754566,
      "balance_loss_mlp": 1.03331614,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.9402259975402543,
      "language_loss": 0.90511495,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92752624,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.691650390625
    },
    {
      "auxiliary_loss_clip": 0.01229275,
      "auxiliary_loss_mlp": 0.00904734,
      "balance_loss_clip": 1.06535852,
      "balance_loss_mlp": 1.00149405,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.608614850245528,
      "language_loss": 0.77630877,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.79764885,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.6221656799316406
    },
    {
      "auxiliary_loss_clip": 0.01216328,
      "auxiliary_loss_mlp": 0.01046599,
      "balance_loss_clip": 1.07050467,
      "balance_loss_mlp": 1.03538132,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.947176841292276,
      "language_loss": 0.85968947,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88231874,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.690690755844116
    },
    {
      "auxiliary_loss_clip": 0.01216787,
      "auxiliary_loss_mlp": 0.01037722,
      "balance_loss_clip": 1.06640553,
      "balance_loss_mlp": 1.02739275,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 2.19509979236143,
      "language_loss": 0.79025102,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81279612,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.642636775970459
    },
    {
      "auxiliary_loss_clip": 0.01206966,
      "auxiliary_loss_mlp": 0.01036353,
      "balance_loss_clip": 1.06400168,
      "balance_loss_mlp": 1.02571988,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 5.128055906309742,
      "language_loss": 0.72276288,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.7451961,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.8516156673431396
    },
    {
      "auxiliary_loss_clip": 0.01225122,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.06521678,
      "balance_loss_mlp": 1.02313852,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 2.270846592096681,
      "language_loss": 0.94330573,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.9658975,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.5841548442840576
    },
    {
      "auxiliary_loss_clip": 0.01174432,
      "auxiliary_loss_mlp": 0.01043309,
      "balance_loss_clip": 1.05698156,
      "balance_loss_mlp": 1.03234208,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.244228951782622,
      "language_loss": 0.88244593,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90462339,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.7813870906829834
    },
    {
      "auxiliary_loss_clip": 0.01204545,
      "auxiliary_loss_mlp": 0.01032896,
      "balance_loss_clip": 1.0619415,
      "balance_loss_mlp": 1.02231598,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 1.9518306049856613,
      "language_loss": 0.89222515,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.9145996,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.827495813369751
    },
    {
      "auxiliary_loss_clip": 0.01198252,
      "auxiliary_loss_mlp": 0.01040915,
      "balance_loss_clip": 1.05959964,
      "balance_loss_mlp": 1.02907157,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 2.1066594017540954,
      "language_loss": 0.88081574,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90320742,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.6565303802490234
    },
    {
      "auxiliary_loss_clip": 0.012108,
      "auxiliary_loss_mlp": 0.01032564,
      "balance_loss_clip": 1.06259668,
      "balance_loss_mlp": 1.02205586,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 2.3410385447148014,
      "language_loss": 0.73143017,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75386381,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.731368064880371
    },
    {
      "auxiliary_loss_clip": 0.01229574,
      "auxiliary_loss_mlp": 0.01039786,
      "balance_loss_clip": 1.07007504,
      "balance_loss_mlp": 1.02850902,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 1.7771029213832759,
      "language_loss": 0.77758312,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.8002767,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.6341328620910645
    },
    {
      "auxiliary_loss_clip": 0.01222024,
      "auxiliary_loss_mlp": 0.01038982,
      "balance_loss_clip": 1.06473255,
      "balance_loss_mlp": 1.0280087,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.8363037246028866,
      "language_loss": 0.8263101,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84892017,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.7196125984191895
    },
    {
      "auxiliary_loss_clip": 0.01212834,
      "auxiliary_loss_mlp": 0.01033738,
      "balance_loss_clip": 1.06455255,
      "balance_loss_mlp": 1.02285361,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 5.359440666901652,
      "language_loss": 0.84923339,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87169909,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.6350204944610596
    },
    {
      "auxiliary_loss_clip": 0.01194196,
      "auxiliary_loss_mlp": 0.0104092,
      "balance_loss_clip": 1.05942798,
      "balance_loss_mlp": 1.02981544,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.9343804874379344,
      "language_loss": 0.76196373,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78431487,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.8405978679656982
    },
    {
      "auxiliary_loss_clip": 0.01221473,
      "auxiliary_loss_mlp": 0.01035167,
      "balance_loss_clip": 1.06146801,
      "balance_loss_mlp": 1.02464628,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 2.0776973790647375,
      "language_loss": 0.74317074,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76573706,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.622124433517456
    },
    {
      "auxiliary_loss_clip": 0.0121717,
      "auxiliary_loss_mlp": 0.01041871,
      "balance_loss_clip": 1.06618702,
      "balance_loss_mlp": 1.03154731,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 2.06008120043034,
      "language_loss": 0.77783334,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.80042374,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 4.461273670196533
    },
    {
      "auxiliary_loss_clip": 0.01226272,
      "auxiliary_loss_mlp": 0.0103902,
      "balance_loss_clip": 1.06979823,
      "balance_loss_mlp": 1.02842259,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.9939162063873719,
      "language_loss": 0.75655055,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.77920347,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.7861876487731934
    },
    {
      "auxiliary_loss_clip": 0.01211173,
      "auxiliary_loss_mlp": 0.01035703,
      "balance_loss_clip": 1.06202316,
      "balance_loss_mlp": 1.02371693,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 2.134582760890667,
      "language_loss": 0.86696994,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.88943869,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 3.693491220474243
    },
    {
      "auxiliary_loss_clip": 0.01230367,
      "auxiliary_loss_mlp": 0.01041838,
      "balance_loss_clip": 1.0708462,
      "balance_loss_mlp": 1.0311631,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 4.795811469285205,
      "language_loss": 0.80199862,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82472074,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 3.671492576599121
    },
    {
      "auxiliary_loss_clip": 0.0121243,
      "auxiliary_loss_mlp": 0.01031629,
      "balance_loss_clip": 1.06708443,
      "balance_loss_mlp": 1.02095354,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 2.172791320773103,
      "language_loss": 0.76524198,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78768259,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.752889394760132
    },
    {
      "auxiliary_loss_clip": 0.01217732,
      "auxiliary_loss_mlp": 0.01046106,
      "balance_loss_clip": 1.06288433,
      "balance_loss_mlp": 1.03429222,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 4.204959740617857,
      "language_loss": 0.85856044,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88119876,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.5891823768615723
    },
    {
      "auxiliary_loss_clip": 0.01205108,
      "auxiliary_loss_mlp": 0.0103447,
      "balance_loss_clip": 1.05947566,
      "balance_loss_mlp": 1.0241046,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 2.0285416315188214,
      "language_loss": 0.82546949,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84786528,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.7070846557617188
    },
    {
      "auxiliary_loss_clip": 0.01205727,
      "auxiliary_loss_mlp": 0.01040805,
      "balance_loss_clip": 1.06162047,
      "balance_loss_mlp": 1.02986109,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 2.9410655705987265,
      "language_loss": 0.79531473,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81778008,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.7558224201202393
    },
    {
      "auxiliary_loss_clip": 0.01136139,
      "auxiliary_loss_mlp": 0.01005009,
      "balance_loss_clip": 1.03956032,
      "balance_loss_mlp": 1.00217235,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8850295091247394,
      "language_loss": 0.59269226,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61410367,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.1666431427001953
    },
    {
      "auxiliary_loss_clip": 0.01223796,
      "auxiliary_loss_mlp": 0.01039775,
      "balance_loss_clip": 1.06605315,
      "balance_loss_mlp": 1.02861071,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 2.0755345841508053,
      "language_loss": 0.74929386,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77192962,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.6589133739471436
    },
    {
      "auxiliary_loss_clip": 0.01214808,
      "auxiliary_loss_mlp": 0.01044272,
      "balance_loss_clip": 1.06857896,
      "balance_loss_mlp": 1.03292954,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 3.154495900145308,
      "language_loss": 0.84690905,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.8694998,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.6422135829925537
    },
    {
      "auxiliary_loss_clip": 0.01232413,
      "auxiliary_loss_mlp": 0.01046342,
      "balance_loss_clip": 1.06733406,
      "balance_loss_mlp": 1.03448057,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.483806812517115,
      "language_loss": 0.62184948,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64463705,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.6701011657714844
    },
    {
      "auxiliary_loss_clip": 0.01217727,
      "auxiliary_loss_mlp": 0.01040204,
      "balance_loss_clip": 1.06557631,
      "balance_loss_mlp": 1.02948141,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.428497850996622,
      "language_loss": 0.86596251,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88854182,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.6613221168518066
    },
    {
      "auxiliary_loss_clip": 0.01197366,
      "auxiliary_loss_mlp": 0.01042177,
      "balance_loss_clip": 1.06000519,
      "balance_loss_mlp": 1.03169203,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 4.240753154796574,
      "language_loss": 0.71617949,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73857486,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.708130359649658
    },
    {
      "auxiliary_loss_clip": 0.01228981,
      "auxiliary_loss_mlp": 0.01039365,
      "balance_loss_clip": 1.0712781,
      "balance_loss_mlp": 1.02907753,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.5418618725163618,
      "language_loss": 0.88163018,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90431362,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.597407102584839
    },
    {
      "auxiliary_loss_clip": 0.01236009,
      "auxiliary_loss_mlp": 0.0103674,
      "balance_loss_clip": 1.06565762,
      "balance_loss_mlp": 1.02524805,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 2.4618798628780856,
      "language_loss": 0.81061339,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83334088,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.64591121673584
    },
    {
      "auxiliary_loss_clip": 0.01225639,
      "auxiliary_loss_mlp": 0.01038506,
      "balance_loss_clip": 1.06723559,
      "balance_loss_mlp": 1.02805746,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 4.376591191729176,
      "language_loss": 0.83264828,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.8552897,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.6067049503326416
    },
    {
      "auxiliary_loss_clip": 0.01207981,
      "auxiliary_loss_mlp": 0.01043093,
      "balance_loss_clip": 1.0607518,
      "balance_loss_mlp": 1.03145838,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.3630303297419806,
      "language_loss": 0.77956641,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80207717,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.6711230278015137
    },
    {
      "auxiliary_loss_clip": 0.01230894,
      "auxiliary_loss_mlp": 0.01034659,
      "balance_loss_clip": 1.06839728,
      "balance_loss_mlp": 1.02394819,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.8396500178485236,
      "language_loss": 0.8099333,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83258891,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.5859391689300537
    },
    {
      "auxiliary_loss_clip": 0.01118465,
      "auxiliary_loss_mlp": 0.01004408,
      "balance_loss_clip": 1.03124142,
      "balance_loss_mlp": 1.00166643,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.81076861906953,
      "language_loss": 0.61909056,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64031923,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.1725711822509766
    },
    {
      "auxiliary_loss_clip": 0.01225931,
      "auxiliary_loss_mlp": 0.01042775,
      "balance_loss_clip": 1.06773889,
      "balance_loss_mlp": 1.02997243,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 3.6260446269477837,
      "language_loss": 0.63746321,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66015029,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.772644519805908
    },
    {
      "auxiliary_loss_clip": 0.01234068,
      "auxiliary_loss_mlp": 0.01044687,
      "balance_loss_clip": 1.06482494,
      "balance_loss_mlp": 1.0332309,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 3.5573541564618725,
      "language_loss": 0.81005752,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83284509,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.632901191711426
    },
    {
      "auxiliary_loss_clip": 0.01194342,
      "auxiliary_loss_mlp": 0.01040286,
      "balance_loss_clip": 1.0544256,
      "balance_loss_mlp": 1.02896702,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.9083903054423303,
      "language_loss": 0.78192198,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80426824,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.7835707664489746
    },
    {
      "auxiliary_loss_clip": 0.01193431,
      "auxiliary_loss_mlp": 0.01051154,
      "balance_loss_clip": 1.05822194,
      "balance_loss_mlp": 1.03999019,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.2703373496133947,
      "language_loss": 0.85413498,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87658083,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.779147148132324
    },
    {
      "auxiliary_loss_clip": 0.01192109,
      "auxiliary_loss_mlp": 0.01044038,
      "balance_loss_clip": 1.05592477,
      "balance_loss_mlp": 1.03220081,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 2.0226890005517797,
      "language_loss": 0.82912236,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.85148382,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.8040120601654053
    },
    {
      "auxiliary_loss_clip": 0.0123936,
      "auxiliary_loss_mlp": 0.01042759,
      "balance_loss_clip": 1.06921732,
      "balance_loss_mlp": 1.03125513,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.396780003367694,
      "language_loss": 0.76571476,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78853595,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 2.60542368888855
    },
    {
      "auxiliary_loss_clip": 0.01207667,
      "auxiliary_loss_mlp": 0.01038796,
      "balance_loss_clip": 1.05878627,
      "balance_loss_mlp": 1.02682781,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 3.3200883854342957,
      "language_loss": 0.73194665,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75441128,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 3.664416551589966
    },
    {
      "auxiliary_loss_clip": 0.0122947,
      "auxiliary_loss_mlp": 0.01037921,
      "balance_loss_clip": 1.06339812,
      "balance_loss_mlp": 1.02638173,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 3.095703708759512,
      "language_loss": 0.77752709,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.80020094,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 3.6179025173187256
    },
    {
      "auxiliary_loss_clip": 0.01227042,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.06548119,
      "balance_loss_mlp": 1.02506173,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 2.5646297575367485,
      "language_loss": 0.85130572,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87393665,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.59415340423584
    },
    {
      "auxiliary_loss_clip": 0.01203713,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.06004,
      "balance_loss_mlp": 1.02507234,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 3.625627244254448,
      "language_loss": 0.9024806,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92487955,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 3.695868730545044
    },
    {
      "auxiliary_loss_clip": 0.01236741,
      "auxiliary_loss_mlp": 0.01035701,
      "balance_loss_clip": 1.06828618,
      "balance_loss_mlp": 1.02515137,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 9.522959891816258,
      "language_loss": 0.7659899,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78871429,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 3.6305198669433594
    },
    {
      "auxiliary_loss_clip": 0.01224918,
      "auxiliary_loss_mlp": 0.01043375,
      "balance_loss_clip": 1.06565309,
      "balance_loss_mlp": 1.03309906,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 5.9856684380625405,
      "language_loss": 0.78717744,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.80986047,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.6269547939300537
    },
    {
      "auxiliary_loss_clip": 0.0121417,
      "auxiliary_loss_mlp": 0.01035931,
      "balance_loss_clip": 1.06360674,
      "balance_loss_mlp": 1.02429032,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.874484127328751,
      "language_loss": 0.82384318,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.84634423,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.7730484008789062
    },
    {
      "auxiliary_loss_clip": 0.01223301,
      "auxiliary_loss_mlp": 0.01044057,
      "balance_loss_clip": 1.06354284,
      "balance_loss_mlp": 1.03300059,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.740406542244589,
      "language_loss": 0.77284122,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.79551482,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.5818467140197754
    },
    {
      "auxiliary_loss_clip": 0.01231949,
      "auxiliary_loss_mlp": 0.01038695,
      "balance_loss_clip": 1.06487083,
      "balance_loss_mlp": 1.02710795,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.571026376113581,
      "language_loss": 0.83882618,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86153257,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.59440279006958
    },
    {
      "auxiliary_loss_clip": 0.01200741,
      "auxiliary_loss_mlp": 0.01045479,
      "balance_loss_clip": 1.06109929,
      "balance_loss_mlp": 1.03413582,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.7913844958571032,
      "language_loss": 0.76705664,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.78951883,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.8504209518432617
    },
    {
      "auxiliary_loss_clip": 0.01200393,
      "auxiliary_loss_mlp": 0.0103857,
      "balance_loss_clip": 1.05773783,
      "balance_loss_mlp": 1.02754283,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.3049215495484394,
      "language_loss": 0.78197283,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80436242,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.8356142044067383
    },
    {
      "auxiliary_loss_clip": 0.01220031,
      "auxiliary_loss_mlp": 0.01034025,
      "balance_loss_clip": 1.06331706,
      "balance_loss_mlp": 1.0230335,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 1.9854989669832903,
      "language_loss": 0.74584931,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.76838982,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.613182306289673
    },
    {
      "auxiliary_loss_clip": 0.0123577,
      "auxiliary_loss_mlp": 0.01034905,
      "balance_loss_clip": 1.06745076,
      "balance_loss_mlp": 1.02384257,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 2.7626567410006158,
      "language_loss": 0.75997448,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78268123,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.6526191234588623
    },
    {
      "auxiliary_loss_clip": 0.01218967,
      "auxiliary_loss_mlp": 0.01048387,
      "balance_loss_clip": 1.06008387,
      "balance_loss_mlp": 1.03679991,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.4329341395416684,
      "language_loss": 0.79876935,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.8214429,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.667433500289917
    },
    {
      "auxiliary_loss_clip": 0.01215491,
      "auxiliary_loss_mlp": 0.01039989,
      "balance_loss_clip": 1.06263256,
      "balance_loss_mlp": 1.02911687,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 3.7329572066486874,
      "language_loss": 0.78576112,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80831587,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.607712507247925
    },
    {
      "auxiliary_loss_clip": 0.01176635,
      "auxiliary_loss_mlp": 0.01042063,
      "balance_loss_clip": 1.05558431,
      "balance_loss_mlp": 1.03110743,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.0320102488554537,
      "language_loss": 0.78464162,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80682856,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.788922071456909
    },
    {
      "auxiliary_loss_clip": 0.01196856,
      "auxiliary_loss_mlp": 0.01038852,
      "balance_loss_clip": 1.05881238,
      "balance_loss_mlp": 1.02704477,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 3.870819849720588,
      "language_loss": 0.65907282,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.68142992,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.692096471786499
    },
    {
      "auxiliary_loss_clip": 0.01210512,
      "auxiliary_loss_mlp": 0.0103909,
      "balance_loss_clip": 1.05873859,
      "balance_loss_mlp": 1.02797365,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 2.0032461881128314,
      "language_loss": 0.89107841,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91357446,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.7102737426757812
    },
    {
      "auxiliary_loss_clip": 0.01235962,
      "auxiliary_loss_mlp": 0.0103618,
      "balance_loss_clip": 1.06788802,
      "balance_loss_mlp": 1.02551699,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.780073016192071,
      "language_loss": 0.78827667,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81099808,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.582254648208618
    },
    {
      "auxiliary_loss_clip": 0.01184355,
      "auxiliary_loss_mlp": 0.01038511,
      "balance_loss_clip": 1.0557518,
      "balance_loss_mlp": 1.0269599,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 4.879784553733243,
      "language_loss": 0.7868917,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.80912036,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.697824716567993
    },
    {
      "auxiliary_loss_clip": 0.01198257,
      "auxiliary_loss_mlp": 0.01037908,
      "balance_loss_clip": 1.062397,
      "balance_loss_mlp": 1.02574885,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 5.392227118499811,
      "language_loss": 0.7387498,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.7611115,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.682110548019409
    },
    {
      "auxiliary_loss_clip": 0.01202818,
      "auxiliary_loss_mlp": 0.01048147,
      "balance_loss_clip": 1.06498063,
      "balance_loss_mlp": 1.03685212,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.592759579452426,
      "language_loss": 0.75269246,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77520216,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.8338944911956787
    },
    {
      "auxiliary_loss_clip": 0.01186507,
      "auxiliary_loss_mlp": 0.00905333,
      "balance_loss_clip": 1.057019,
      "balance_loss_mlp": 1.0020237,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.112346464632047,
      "language_loss": 0.76911342,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.79003185,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.6596341133117676
    },
    {
      "auxiliary_loss_clip": 0.0122278,
      "auxiliary_loss_mlp": 0.01039656,
      "balance_loss_clip": 1.06578207,
      "balance_loss_mlp": 1.02887368,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.534329275811449,
      "language_loss": 0.71074486,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73336923,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.670886516571045
    },
    {
      "auxiliary_loss_clip": 0.01228616,
      "auxiliary_loss_mlp": 0.01041308,
      "balance_loss_clip": 1.0652529,
      "balance_loss_mlp": 1.03013206,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.6243966298622494,
      "language_loss": 0.75990891,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78260815,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.654787540435791
    },
    {
      "auxiliary_loss_clip": 0.01192925,
      "auxiliary_loss_mlp": 0.01042587,
      "balance_loss_clip": 1.05367827,
      "balance_loss_mlp": 1.03098178,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 3.138787615732323,
      "language_loss": 0.80642301,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.82877815,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.7040321826934814
    },
    {
      "auxiliary_loss_clip": 0.01229962,
      "auxiliary_loss_mlp": 0.01033335,
      "balance_loss_clip": 1.06366539,
      "balance_loss_mlp": 1.02319038,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.1714964004193127,
      "language_loss": 0.78085726,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80349028,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.592578649520874
    },
    {
      "auxiliary_loss_clip": 0.01209443,
      "auxiliary_loss_mlp": 0.01037927,
      "balance_loss_clip": 1.06637096,
      "balance_loss_mlp": 1.02766895,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 3.1924693906104555,
      "language_loss": 0.75350797,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.77598166,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 3.576671838760376
    },
    {
      "auxiliary_loss_clip": 0.01189514,
      "auxiliary_loss_mlp": 0.01038931,
      "balance_loss_clip": 1.0522418,
      "balance_loss_mlp": 1.02824426,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 2.0231119474349364,
      "language_loss": 0.80169499,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82397944,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 3.695502281188965
    },
    {
      "auxiliary_loss_clip": 0.01203953,
      "auxiliary_loss_mlp": 0.01037336,
      "balance_loss_clip": 1.05772257,
      "balance_loss_mlp": 1.02665496,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.4198265237120697,
      "language_loss": 0.89361411,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91602695,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 3.565275192260742
    },
    {
      "auxiliary_loss_clip": 0.0123046,
      "auxiliary_loss_mlp": 0.01044692,
      "balance_loss_clip": 1.067132,
      "balance_loss_mlp": 1.03321195,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 3.103802927400531,
      "language_loss": 0.82010847,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84285998,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 3.527714967727661
    },
    {
      "auxiliary_loss_clip": 0.01228258,
      "auxiliary_loss_mlp": 0.00905488,
      "balance_loss_clip": 1.06399727,
      "balance_loss_mlp": 1.00172687,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 2.1448171778336054,
      "language_loss": 0.79831493,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.81965244,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.603132963180542
    },
    {
      "auxiliary_loss_clip": 0.01207589,
      "auxiliary_loss_mlp": 0.01039662,
      "balance_loss_clip": 1.06297708,
      "balance_loss_mlp": 1.02856326,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.443800147716788,
      "language_loss": 0.75784856,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78032106,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.726144790649414
    },
    {
      "auxiliary_loss_clip": 0.01222661,
      "auxiliary_loss_mlp": 0.01045021,
      "balance_loss_clip": 1.06711602,
      "balance_loss_mlp": 1.0345186,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 2.0536953865138865,
      "language_loss": 0.85027152,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87294835,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.6471619606018066
    },
    {
      "auxiliary_loss_clip": 0.01215564,
      "auxiliary_loss_mlp": 0.0103795,
      "balance_loss_clip": 1.05921078,
      "balance_loss_mlp": 1.02708447,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.9893476794618583,
      "language_loss": 0.81200695,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83454216,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.7177014350891113
    },
    {
      "auxiliary_loss_clip": 0.01188995,
      "auxiliary_loss_mlp": 0.01031007,
      "balance_loss_clip": 1.0553329,
      "balance_loss_mlp": 1.01937795,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.433626520683159,
      "language_loss": 0.7748189,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79701889,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.728332996368408
    },
    {
      "auxiliary_loss_clip": 0.01228556,
      "auxiliary_loss_mlp": 0.01047068,
      "balance_loss_clip": 1.06543636,
      "balance_loss_mlp": 1.03548717,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 3.031960447780468,
      "language_loss": 0.88807702,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.91083324,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.7040514945983887
    },
    {
      "auxiliary_loss_clip": 0.01206073,
      "auxiliary_loss_mlp": 0.01033965,
      "balance_loss_clip": 1.05650401,
      "balance_loss_mlp": 1.02328372,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.739729962233288,
      "language_loss": 0.7034601,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.72586048,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.6623919010162354
    },
    {
      "auxiliary_loss_clip": 0.01231327,
      "auxiliary_loss_mlp": 0.00904363,
      "balance_loss_clip": 1.06791377,
      "balance_loss_mlp": 1.00156176,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.035492342273551,
      "language_loss": 0.88937676,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.9107337,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.689624547958374
    },
    {
      "auxiliary_loss_clip": 0.01214906,
      "auxiliary_loss_mlp": 0.01041591,
      "balance_loss_clip": 1.06191897,
      "balance_loss_mlp": 1.02981317,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.7348897446772769,
      "language_loss": 0.81790841,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.84047341,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.67159366607666
    },
    {
      "auxiliary_loss_clip": 0.01205384,
      "auxiliary_loss_mlp": 0.01036091,
      "balance_loss_clip": 1.06016207,
      "balance_loss_mlp": 1.02600598,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 1.9362863984947924,
      "language_loss": 0.81125629,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83367109,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.7010908126831055
    },
    {
      "auxiliary_loss_clip": 0.0122417,
      "auxiliary_loss_mlp": 0.01032267,
      "balance_loss_clip": 1.06436729,
      "balance_loss_mlp": 1.02255702,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 1.907969837734579,
      "language_loss": 0.79556096,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81812525,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.6220200061798096
    },
    {
      "auxiliary_loss_clip": 0.01237153,
      "auxiliary_loss_mlp": 0.01045918,
      "balance_loss_clip": 1.06911397,
      "balance_loss_mlp": 1.03516567,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 3.2376368982219352,
      "language_loss": 0.79827833,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82110906,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.582033395767212
    },
    {
      "auxiliary_loss_clip": 0.01210061,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.06094182,
      "balance_loss_mlp": 1.02000129,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.419108316518698,
      "language_loss": 0.69063079,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71304172,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.5976648330688477
    },
    {
      "auxiliary_loss_clip": 0.01198279,
      "auxiliary_loss_mlp": 0.01037657,
      "balance_loss_clip": 1.05882883,
      "balance_loss_mlp": 1.0270716,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 4.743200230139598,
      "language_loss": 0.94876695,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.97112632,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.79034161567688
    },
    {
      "auxiliary_loss_clip": 0.01235605,
      "auxiliary_loss_mlp": 0.01041458,
      "balance_loss_clip": 1.06609952,
      "balance_loss_mlp": 1.03059244,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 3.6357678423703215,
      "language_loss": 0.82013822,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84290886,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.53180193901062
    },
    {
      "auxiliary_loss_clip": 0.01214022,
      "auxiliary_loss_mlp": 0.01046068,
      "balance_loss_clip": 1.06017959,
      "balance_loss_mlp": 1.03554153,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 2.089515448574186,
      "language_loss": 0.67246675,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69506764,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.793708562850952
    },
    {
      "auxiliary_loss_clip": 0.01185398,
      "auxiliary_loss_mlp": 0.01039266,
      "balance_loss_clip": 1.05622315,
      "balance_loss_mlp": 1.02859116,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.9160555829085728,
      "language_loss": 0.7295956,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75184226,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.8773515224456787
    },
    {
      "auxiliary_loss_clip": 0.01221283,
      "auxiliary_loss_mlp": 0.01036206,
      "balance_loss_clip": 1.06992066,
      "balance_loss_mlp": 1.02620411,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.273306585812516,
      "language_loss": 0.78009903,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80267394,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.5936379432678223
    },
    {
      "auxiliary_loss_clip": 0.0118548,
      "auxiliary_loss_mlp": 0.00904607,
      "balance_loss_clip": 1.06206942,
      "balance_loss_mlp": 1.00165713,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.824627455381715,
      "language_loss": 0.79622763,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81712854,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.7543187141418457
    },
    {
      "auxiliary_loss_clip": 0.01213402,
      "auxiliary_loss_mlp": 0.01037612,
      "balance_loss_clip": 1.06438172,
      "balance_loss_mlp": 1.02669191,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 6.505515345557996,
      "language_loss": 0.88869131,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91120148,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.681830644607544
    },
    {
      "auxiliary_loss_clip": 0.01236346,
      "auxiliary_loss_mlp": 0.01038972,
      "balance_loss_clip": 1.07174659,
      "balance_loss_mlp": 1.0284462,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 2.993915207039921,
      "language_loss": 0.80298114,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82573438,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.6150355339050293
    },
    {
      "auxiliary_loss_clip": 0.01232216,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.07034421,
      "balance_loss_mlp": 1.02809989,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 2.8127525546363534,
      "language_loss": 0.81090736,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.8336333,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.6209805011749268
    },
    {
      "auxiliary_loss_clip": 0.01190876,
      "auxiliary_loss_mlp": 0.01041129,
      "balance_loss_clip": 1.05868077,
      "balance_loss_mlp": 1.0291543,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 1.814449243797795,
      "language_loss": 0.6430459,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66536593,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.7685635089874268
    },
    {
      "auxiliary_loss_clip": 0.01198062,
      "auxiliary_loss_mlp": 0.01044386,
      "balance_loss_clip": 1.05955839,
      "balance_loss_mlp": 1.03357959,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.9699362778562444,
      "language_loss": 0.80689573,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82932019,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 4.577579975128174
    },
    {
      "auxiliary_loss_clip": 0.01203602,
      "auxiliary_loss_mlp": 0.01037532,
      "balance_loss_clip": 1.06688535,
      "balance_loss_mlp": 1.02605784,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.7121570004574194,
      "language_loss": 0.83074617,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85315752,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.687817096710205
    },
    {
      "auxiliary_loss_clip": 0.0123763,
      "auxiliary_loss_mlp": 0.01038205,
      "balance_loss_clip": 1.06918657,
      "balance_loss_mlp": 1.02697563,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.8643509679578054,
      "language_loss": 0.51726723,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.54002559,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 3.552328109741211
    },
    {
      "auxiliary_loss_clip": 0.01213356,
      "auxiliary_loss_mlp": 0.01036581,
      "balance_loss_clip": 1.06582403,
      "balance_loss_mlp": 1.02409983,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 2.2649093723501705,
      "language_loss": 0.77916664,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80166602,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 3.6197612285614014
    },
    {
      "auxiliary_loss_clip": 0.01213598,
      "auxiliary_loss_mlp": 0.0103478,
      "balance_loss_clip": 1.06095254,
      "balance_loss_mlp": 1.02268052,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 6.652940539852674,
      "language_loss": 0.72759867,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.75008243,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.660250425338745
    },
    {
      "auxiliary_loss_clip": 0.01241111,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 1.07351279,
      "balance_loss_mlp": 1.03051543,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 3.2128701908651767,
      "language_loss": 0.81820369,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84102762,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.630983829498291
    },
    {
      "auxiliary_loss_clip": 0.01206172,
      "auxiliary_loss_mlp": 0.01037746,
      "balance_loss_clip": 1.06373811,
      "balance_loss_mlp": 1.0265578,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 2.082654614588465,
      "language_loss": 0.69522429,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71766347,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.7465262413024902
    },
    {
      "auxiliary_loss_clip": 0.01237124,
      "auxiliary_loss_mlp": 0.01038525,
      "balance_loss_clip": 1.07027769,
      "balance_loss_mlp": 1.02867794,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.3976522481433284,
      "language_loss": 0.91560566,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.93836212,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.6345956325531006
    },
    {
      "auxiliary_loss_clip": 0.01228889,
      "auxiliary_loss_mlp": 0.01039702,
      "balance_loss_clip": 1.0713129,
      "balance_loss_mlp": 1.02896738,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 2.232794926134513,
      "language_loss": 0.88613772,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90882373,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.6609249114990234
    },
    {
      "auxiliary_loss_clip": 0.01222135,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.06576538,
      "balance_loss_mlp": 1.02186584,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 1.929251966031264,
      "language_loss": 0.75800276,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.7805562,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.7458298206329346
    },
    {
      "auxiliary_loss_clip": 0.01217322,
      "auxiliary_loss_mlp": 0.0104466,
      "balance_loss_clip": 1.06818306,
      "balance_loss_mlp": 1.03360975,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.6887264692838642,
      "language_loss": 0.79068542,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81330526,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.672039270401001
    },
    {
      "auxiliary_loss_clip": 0.01224646,
      "auxiliary_loss_mlp": 0.01037143,
      "balance_loss_clip": 1.06919074,
      "balance_loss_mlp": 1.02631283,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 1.7956396585116918,
      "language_loss": 0.84799069,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87060857,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.6404995918273926
    },
    {
      "auxiliary_loss_clip": 0.01221951,
      "auxiliary_loss_mlp": 0.0103663,
      "balance_loss_clip": 1.06940866,
      "balance_loss_mlp": 1.02451801,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.666226979162642,
      "language_loss": 0.71619797,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73878378,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.623110055923462
    },
    {
      "auxiliary_loss_clip": 0.01161985,
      "auxiliary_loss_mlp": 0.01042953,
      "balance_loss_clip": 1.05035281,
      "balance_loss_mlp": 1.03221214,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 6.413557302341085,
      "language_loss": 0.78833002,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.81037939,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.7945659160614014
    },
    {
      "auxiliary_loss_clip": 0.01183334,
      "auxiliary_loss_mlp": 0.01047226,
      "balance_loss_clip": 1.06384957,
      "balance_loss_mlp": 1.0370934,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.3376097038219275,
      "language_loss": 0.7560215,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.77832717,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.7205898761749268
    },
    {
      "auxiliary_loss_clip": 0.01228554,
      "auxiliary_loss_mlp": 0.0103983,
      "balance_loss_clip": 1.07105231,
      "balance_loss_mlp": 1.02976859,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 2.3156331339702176,
      "language_loss": 0.71469712,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73738098,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.6785101890563965
    },
    {
      "auxiliary_loss_clip": 0.01215824,
      "auxiliary_loss_mlp": 0.01048695,
      "balance_loss_clip": 1.06522167,
      "balance_loss_mlp": 1.03720355,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.9320133841473275,
      "language_loss": 0.73419726,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75684243,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.6880054473876953
    },
    {
      "auxiliary_loss_clip": 0.01173428,
      "auxiliary_loss_mlp": 0.01044793,
      "balance_loss_clip": 1.05354071,
      "balance_loss_mlp": 1.03324163,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 2.040509474030294,
      "language_loss": 0.76479727,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.7869795,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.702622890472412
    },
    {
      "auxiliary_loss_clip": 0.01232887,
      "auxiliary_loss_mlp": 0.00904626,
      "balance_loss_clip": 1.0686146,
      "balance_loss_mlp": 1.00173676,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.398517995642926,
      "language_loss": 0.84742957,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86880469,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.6636803150177
    },
    {
      "auxiliary_loss_clip": 0.01213287,
      "auxiliary_loss_mlp": 0.01038577,
      "balance_loss_clip": 1.06496286,
      "balance_loss_mlp": 1.02819967,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.8144111697012604,
      "language_loss": 0.84967852,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87219715,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.6459903717041016
    },
    {
      "auxiliary_loss_clip": 0.01204805,
      "auxiliary_loss_mlp": 0.01040012,
      "balance_loss_clip": 1.05975676,
      "balance_loss_mlp": 1.02831721,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 3.0759643071524962,
      "language_loss": 0.72769499,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75014311,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.653829336166382
    },
    {
      "auxiliary_loss_clip": 0.01112125,
      "auxiliary_loss_mlp": 0.01009244,
      "balance_loss_clip": 1.02902746,
      "balance_loss_mlp": 1.00602555,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.713852638868898,
      "language_loss": 0.54465997,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56587368,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.36979079246521
    },
    {
      "auxiliary_loss_clip": 0.01239694,
      "auxiliary_loss_mlp": 0.0103384,
      "balance_loss_clip": 1.07129121,
      "balance_loss_mlp": 1.02317095,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 2.194999107738421,
      "language_loss": 0.70624292,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.72897828,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.5968408584594727
    },
    {
      "auxiliary_loss_clip": 0.01236417,
      "auxiliary_loss_mlp": 0.01041377,
      "balance_loss_clip": 1.07007515,
      "balance_loss_mlp": 1.03072596,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.394927866344042,
      "language_loss": 0.78459561,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80737358,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.6525607109069824
    },
    {
      "auxiliary_loss_clip": 0.01236838,
      "auxiliary_loss_mlp": 0.01040853,
      "balance_loss_clip": 1.07167315,
      "balance_loss_mlp": 1.02968287,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 3.3256867870112887,
      "language_loss": 0.73754746,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76032436,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.62068510055542
    },
    {
      "auxiliary_loss_clip": 0.01210939,
      "auxiliary_loss_mlp": 0.01037361,
      "balance_loss_clip": 1.07090163,
      "balance_loss_mlp": 1.02528477,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 3.420627642174105,
      "language_loss": 0.73040318,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75288618,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.811687469482422
    },
    {
      "auxiliary_loss_clip": 0.01229425,
      "auxiliary_loss_mlp": 0.01040994,
      "balance_loss_clip": 1.06791639,
      "balance_loss_mlp": 1.03017592,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.109547305750801,
      "language_loss": 0.77243584,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79514003,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 4.554483890533447
    },
    {
      "auxiliary_loss_clip": 0.01216011,
      "auxiliary_loss_mlp": 0.01042342,
      "balance_loss_clip": 1.06297231,
      "balance_loss_mlp": 1.03067732,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 2.106355871092273,
      "language_loss": 0.85500538,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87758887,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.649449586868286
    },
    {
      "auxiliary_loss_clip": 0.012019,
      "auxiliary_loss_mlp": 0.01038798,
      "balance_loss_clip": 1.06710863,
      "balance_loss_mlp": 1.02814639,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.8653714359927087,
      "language_loss": 0.83399516,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85640216,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 3.6482889652252197
    },
    {
      "auxiliary_loss_clip": 0.01210889,
      "auxiliary_loss_mlp": 0.01044133,
      "balance_loss_clip": 1.06342888,
      "balance_loss_mlp": 1.03315973,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 6.9405824510319025,
      "language_loss": 0.610955,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63350523,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.6742773056030273
    },
    {
      "auxiliary_loss_clip": 0.01216302,
      "auxiliary_loss_mlp": 0.01050096,
      "balance_loss_clip": 1.06743264,
      "balance_loss_mlp": 1.038378,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 3.0423440406475386,
      "language_loss": 0.77049774,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79316163,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 3.5057849884033203
    },
    {
      "auxiliary_loss_clip": 0.01204287,
      "auxiliary_loss_mlp": 0.00905839,
      "balance_loss_clip": 1.060655,
      "balance_loss_mlp": 1.00203562,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.398524704431979,
      "language_loss": 0.85242844,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87352973,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.6928293704986572
    },
    {
      "auxiliary_loss_clip": 0.01196447,
      "auxiliary_loss_mlp": 0.01037866,
      "balance_loss_clip": 1.06120121,
      "balance_loss_mlp": 1.02691686,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.5056809290316218,
      "language_loss": 0.76711249,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.78945553,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.708125114440918
    },
    {
      "auxiliary_loss_clip": 0.01210926,
      "auxiliary_loss_mlp": 0.0103961,
      "balance_loss_clip": 1.06601012,
      "balance_loss_mlp": 1.02805865,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 2.079485170292789,
      "language_loss": 0.80027223,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82277763,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.7463159561157227
    },
    {
      "auxiliary_loss_clip": 0.01206638,
      "auxiliary_loss_mlp": 0.01041238,
      "balance_loss_clip": 1.06544065,
      "balance_loss_mlp": 1.03043151,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 2.034357333819168,
      "language_loss": 0.84253955,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86501831,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.6634154319763184
    },
    {
      "auxiliary_loss_clip": 0.01217344,
      "auxiliary_loss_mlp": 0.01036589,
      "balance_loss_clip": 1.06229079,
      "balance_loss_mlp": 1.02553201,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.83604883078462,
      "language_loss": 0.76815748,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.79069686,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.6152384281158447
    },
    {
      "auxiliary_loss_clip": 0.01209252,
      "auxiliary_loss_mlp": 0.01042836,
      "balance_loss_clip": 1.06463861,
      "balance_loss_mlp": 1.03186321,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 2.654700755807087,
      "language_loss": 0.80816799,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83068889,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.66363525390625
    },
    {
      "auxiliary_loss_clip": 0.01209946,
      "auxiliary_loss_mlp": 0.01038008,
      "balance_loss_clip": 1.06593132,
      "balance_loss_mlp": 1.02685046,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 2.3511444805848822,
      "language_loss": 0.86759764,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89007723,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.6729421615600586
    },
    {
      "auxiliary_loss_clip": 0.01234146,
      "auxiliary_loss_mlp": 0.01034083,
      "balance_loss_clip": 1.06717157,
      "balance_loss_mlp": 1.02364659,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 3.6126579635932425,
      "language_loss": 0.76987714,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79255944,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.6924593448638916
    },
    {
      "auxiliary_loss_clip": 0.01196692,
      "auxiliary_loss_mlp": 0.01038138,
      "balance_loss_clip": 1.06313419,
      "balance_loss_mlp": 1.02734935,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.3646370867003985,
      "language_loss": 0.77450806,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79685634,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.68383526802063
    },
    {
      "auxiliary_loss_clip": 0.01204086,
      "auxiliary_loss_mlp": 0.01038867,
      "balance_loss_clip": 1.05732739,
      "balance_loss_mlp": 1.02767324,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.7589301619519304,
      "language_loss": 0.79802322,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.82045281,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.6854007244110107
    },
    {
      "auxiliary_loss_clip": 0.01218286,
      "auxiliary_loss_mlp": 0.01040152,
      "balance_loss_clip": 1.06378305,
      "balance_loss_mlp": 1.02907133,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 2.5119561862299125,
      "language_loss": 0.69758999,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72017443,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.712113857269287
    },
    {
      "auxiliary_loss_clip": 0.01105894,
      "auxiliary_loss_mlp": 0.01009702,
      "balance_loss_clip": 1.02734113,
      "balance_loss_mlp": 1.00595903,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8443252107556668,
      "language_loss": 0.58370447,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60486042,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 2.9917335510253906
    },
    {
      "auxiliary_loss_clip": 0.01186154,
      "auxiliary_loss_mlp": 0.01036601,
      "balance_loss_clip": 1.05738401,
      "balance_loss_mlp": 1.02561021,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 3.728913799348606,
      "language_loss": 0.76669914,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.78892666,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.7221102714538574
    },
    {
      "auxiliary_loss_clip": 0.01193495,
      "auxiliary_loss_mlp": 0.01035529,
      "balance_loss_clip": 1.05939794,
      "balance_loss_mlp": 1.02421653,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 2.3450830486323215,
      "language_loss": 0.85624611,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87853634,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.7104415893554688
    },
    {
      "auxiliary_loss_clip": 0.01218515,
      "auxiliary_loss_mlp": 0.01036749,
      "balance_loss_clip": 1.06231952,
      "balance_loss_mlp": 1.02616286,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.157524022017406,
      "language_loss": 0.80688781,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82944047,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.560086727142334
    },
    {
      "auxiliary_loss_clip": 0.01219285,
      "auxiliary_loss_mlp": 0.01038697,
      "balance_loss_clip": 1.06245399,
      "balance_loss_mlp": 1.02677035,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 3.3410795483427185,
      "language_loss": 0.81965351,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.8422333,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.6368889808654785
    },
    {
      "auxiliary_loss_clip": 0.01228484,
      "auxiliary_loss_mlp": 0.01044992,
      "balance_loss_clip": 1.0694809,
      "balance_loss_mlp": 1.03369701,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 2.068733906437909,
      "language_loss": 0.79258883,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81532359,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.618839979171753
    },
    {
      "auxiliary_loss_clip": 0.01216607,
      "auxiliary_loss_mlp": 0.01046901,
      "balance_loss_clip": 1.06261182,
      "balance_loss_mlp": 1.03621984,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.2405303883519685,
      "language_loss": 0.78342402,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80605912,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.6326944828033447
    },
    {
      "auxiliary_loss_clip": 0.01222823,
      "auxiliary_loss_mlp": 0.0103165,
      "balance_loss_clip": 1.06530237,
      "balance_loss_mlp": 1.02086723,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 2.0036652306359293,
      "language_loss": 0.81300128,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83554602,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.6110599040985107
    },
    {
      "auxiliary_loss_clip": 0.01232288,
      "auxiliary_loss_mlp": 0.01036986,
      "balance_loss_clip": 1.0665288,
      "balance_loss_mlp": 1.02591133,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 8.44351638966767,
      "language_loss": 0.94257021,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.96526301,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.56615948677063
    },
    {
      "auxiliary_loss_clip": 0.0119641,
      "auxiliary_loss_mlp": 0.01044435,
      "balance_loss_clip": 1.06033432,
      "balance_loss_mlp": 1.03338408,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 4.097525271197601,
      "language_loss": 0.62356609,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64597458,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.7560043334960938
    },
    {
      "auxiliary_loss_clip": 0.01118645,
      "auxiliary_loss_mlp": 0.01000661,
      "balance_loss_clip": 1.03121352,
      "balance_loss_mlp": 0.99718016,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.863332662072469,
      "language_loss": 0.68289429,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70408744,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 4.306468486785889
    },
    {
      "auxiliary_loss_clip": 0.01206536,
      "auxiliary_loss_mlp": 0.01040897,
      "balance_loss_clip": 1.06531763,
      "balance_loss_mlp": 1.0302099,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.9396306016493101,
      "language_loss": 0.78447598,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.80695033,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 3.5765788555145264
    },
    {
      "auxiliary_loss_clip": 0.01201807,
      "auxiliary_loss_mlp": 0.01041178,
      "balance_loss_clip": 1.05724859,
      "balance_loss_mlp": 1.03004408,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.8221147936518425,
      "language_loss": 0.88068509,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90311491,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.6359777450561523
    },
    {
      "auxiliary_loss_clip": 0.01205981,
      "auxiliary_loss_mlp": 0.01041846,
      "balance_loss_clip": 1.05890608,
      "balance_loss_mlp": 1.03072369,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.0621479451542544,
      "language_loss": 0.92618304,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.94866139,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 3.575634002685547
    },
    {
      "auxiliary_loss_clip": 0.0119726,
      "auxiliary_loss_mlp": 0.01032707,
      "balance_loss_clip": 1.06065774,
      "balance_loss_mlp": 1.02120936,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 3.893176632339616,
      "language_loss": 0.89074713,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91304684,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 3.5720040798187256
    },
    {
      "auxiliary_loss_clip": 0.01231624,
      "auxiliary_loss_mlp": 0.0090501,
      "balance_loss_clip": 1.06699681,
      "balance_loss_mlp": 1.00204408,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 3.714325368294506,
      "language_loss": 0.75384158,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.775208,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.5827038288116455
    },
    {
      "auxiliary_loss_clip": 0.01202254,
      "auxiliary_loss_mlp": 0.01034321,
      "balance_loss_clip": 1.06219172,
      "balance_loss_mlp": 1.02362823,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.325179594140161,
      "language_loss": 0.77840853,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80077434,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.6486783027648926
    },
    {
      "auxiliary_loss_clip": 0.01110414,
      "auxiliary_loss_mlp": 0.01004309,
      "balance_loss_clip": 1.02564597,
      "balance_loss_mlp": 1.00092399,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.8896864830929958,
      "language_loss": 0.56205761,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58320487,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.257551431655884
    },
    {
      "auxiliary_loss_clip": 0.01234935,
      "auxiliary_loss_mlp": 0.00904395,
      "balance_loss_clip": 1.06878948,
      "balance_loss_mlp": 1.00182652,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 2.899529236299656,
      "language_loss": 0.88869393,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.91008723,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.6078574657440186
    },
    {
      "auxiliary_loss_clip": 0.01235449,
      "auxiliary_loss_mlp": 0.0103822,
      "balance_loss_clip": 1.06642342,
      "balance_loss_mlp": 1.02629912,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 3.18858617272287,
      "language_loss": 0.71027213,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73300886,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.6277852058410645
    },
    {
      "auxiliary_loss_clip": 0.01213272,
      "auxiliary_loss_mlp": 0.01039664,
      "balance_loss_clip": 1.06539273,
      "balance_loss_mlp": 1.02944207,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 27.544341032615687,
      "language_loss": 0.76396143,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.7864908,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.7611987590789795
    },
    {
      "auxiliary_loss_clip": 0.01214969,
      "auxiliary_loss_mlp": 0.01037119,
      "balance_loss_clip": 1.06304193,
      "balance_loss_mlp": 1.02594876,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.4519922875196842,
      "language_loss": 0.82656866,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.8490895,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.690018653869629
    },
    {
      "auxiliary_loss_clip": 0.01210371,
      "auxiliary_loss_mlp": 0.0103773,
      "balance_loss_clip": 1.06585097,
      "balance_loss_mlp": 1.02622068,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.3581514615532884,
      "language_loss": 0.80638671,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82886773,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.5950653553009033
    },
    {
      "auxiliary_loss_clip": 0.01185204,
      "auxiliary_loss_mlp": 0.01036405,
      "balance_loss_clip": 1.05711746,
      "balance_loss_mlp": 1.02449059,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 3.0990162593467896,
      "language_loss": 0.71390486,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73612094,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.8943777084350586
    },
    {
      "auxiliary_loss_clip": 0.01224531,
      "auxiliary_loss_mlp": 0.01038765,
      "balance_loss_clip": 1.06923389,
      "balance_loss_mlp": 1.02773178,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 3.2474441976559167,
      "language_loss": 0.77119881,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79383177,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.614065647125244
    },
    {
      "auxiliary_loss_clip": 0.01234183,
      "auxiliary_loss_mlp": 0.01047458,
      "balance_loss_clip": 1.06684446,
      "balance_loss_mlp": 1.0352509,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 3.6447637510038176,
      "language_loss": 0.78841436,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.81123078,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.6662275791168213
    },
    {
      "auxiliary_loss_clip": 0.01211022,
      "auxiliary_loss_mlp": 0.01038165,
      "balance_loss_clip": 1.0659548,
      "balance_loss_mlp": 1.02709603,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.7451795633505907,
      "language_loss": 0.84413707,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86662894,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.739426851272583
    },
    {
      "auxiliary_loss_clip": 0.01236193,
      "auxiliary_loss_mlp": 0.01036894,
      "balance_loss_clip": 1.06820452,
      "balance_loss_mlp": 1.02558088,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.447495452372098,
      "language_loss": 0.74807417,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.770805,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.6075525283813477
    },
    {
      "auxiliary_loss_clip": 0.0122944,
      "auxiliary_loss_mlp": 0.01037332,
      "balance_loss_clip": 1.06744158,
      "balance_loss_mlp": 1.0252378,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.8079010423668453,
      "language_loss": 0.80639601,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.82906365,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.6316416263580322
    },
    {
      "auxiliary_loss_clip": 0.01205145,
      "auxiliary_loss_mlp": 0.01041627,
      "balance_loss_clip": 1.06305313,
      "balance_loss_mlp": 1.03100514,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 2.205772666637695,
      "language_loss": 0.88189918,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.90436691,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.7855277061462402
    },
    {
      "auxiliary_loss_clip": 0.01218346,
      "auxiliary_loss_mlp": 0.01040078,
      "balance_loss_clip": 1.07119203,
      "balance_loss_mlp": 1.02859831,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 2.2248199238209425,
      "language_loss": 0.83555663,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85814089,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.646684408187866
    },
    {
      "auxiliary_loss_clip": 0.01216945,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.06502652,
      "balance_loss_mlp": 1.02331948,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.1665731585691157,
      "language_loss": 0.79968095,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.82218945,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.532170295715332
    },
    {
      "auxiliary_loss_clip": 0.01232495,
      "auxiliary_loss_mlp": 0.01039515,
      "balance_loss_clip": 1.06868386,
      "balance_loss_mlp": 1.02774334,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 1.9303213747005117,
      "language_loss": 0.72888637,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75160652,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.5920984745025635
    },
    {
      "auxiliary_loss_clip": 0.01204569,
      "auxiliary_loss_mlp": 0.01039467,
      "balance_loss_clip": 1.06199193,
      "balance_loss_mlp": 1.02859545,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 2.0594524408142396,
      "language_loss": 0.69629681,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71873713,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.6397929191589355
    },
    {
      "auxiliary_loss_clip": 0.01225386,
      "auxiliary_loss_mlp": 0.01035977,
      "balance_loss_clip": 1.06434822,
      "balance_loss_mlp": 1.0241394,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 6.076680362745705,
      "language_loss": 0.86712939,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.88974309,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.6044249534606934
    },
    {
      "auxiliary_loss_clip": 0.01208067,
      "auxiliary_loss_mlp": 0.01031457,
      "balance_loss_clip": 1.05908525,
      "balance_loss_mlp": 1.02124667,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 1.9067680609280622,
      "language_loss": 0.79658061,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81897587,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.6660423278808594
    },
    {
      "auxiliary_loss_clip": 0.01187119,
      "auxiliary_loss_mlp": 0.01035809,
      "balance_loss_clip": 1.05755138,
      "balance_loss_mlp": 1.0244422,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.1307898183510687,
      "language_loss": 0.86962581,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89185506,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.715257406234741
    },
    {
      "auxiliary_loss_clip": 0.01171445,
      "auxiliary_loss_mlp": 0.01035647,
      "balance_loss_clip": 1.05581868,
      "balance_loss_mlp": 1.02472711,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 1.7865903263179983,
      "language_loss": 0.70197058,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72404152,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 4.654429197311401
    },
    {
      "auxiliary_loss_clip": 0.01234567,
      "auxiliary_loss_mlp": 0.01044347,
      "balance_loss_clip": 1.0679034,
      "balance_loss_mlp": 1.03399324,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.4030352393218988,
      "language_loss": 0.77129686,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79408598,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.557616949081421
    },
    {
      "auxiliary_loss_clip": 0.01208051,
      "auxiliary_loss_mlp": 0.01033697,
      "balance_loss_clip": 1.06223691,
      "balance_loss_mlp": 1.02200818,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 2.480825281386497,
      "language_loss": 0.69118625,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71360373,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.7375295162200928
    },
    {
      "auxiliary_loss_clip": 0.01219746,
      "auxiliary_loss_mlp": 0.01035231,
      "balance_loss_clip": 1.06555128,
      "balance_loss_mlp": 1.02377534,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.770933255838036,
      "language_loss": 0.83688885,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.85943866,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 3.5759925842285156
    },
    {
      "auxiliary_loss_clip": 0.01235579,
      "auxiliary_loss_mlp": 0.01041732,
      "balance_loss_clip": 1.06764126,
      "balance_loss_mlp": 1.03141427,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.0918831684963894,
      "language_loss": 0.85344183,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87621492,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.5970191955566406
    },
    {
      "auxiliary_loss_clip": 0.01216558,
      "auxiliary_loss_mlp": 0.01041032,
      "balance_loss_clip": 1.06623387,
      "balance_loss_mlp": 1.02927792,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 3.939193512450443,
      "language_loss": 0.83214617,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.85472202,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.6654927730560303
    },
    {
      "auxiliary_loss_clip": 0.01203312,
      "auxiliary_loss_mlp": 0.01046701,
      "balance_loss_clip": 1.06399512,
      "balance_loss_mlp": 1.03551877,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.725784467966601,
      "language_loss": 0.88028377,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90278387,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.7450332641601562
    },
    {
      "auxiliary_loss_clip": 0.0121117,
      "auxiliary_loss_mlp": 0.01040439,
      "balance_loss_clip": 1.06481194,
      "balance_loss_mlp": 1.02966261,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 2.1556585663594183,
      "language_loss": 0.7912389,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81375504,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.6565747261047363
    },
    {
      "auxiliary_loss_clip": 0.01193453,
      "auxiliary_loss_mlp": 0.01034014,
      "balance_loss_clip": 1.05711436,
      "balance_loss_mlp": 1.02211726,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 3.400127505510758,
      "language_loss": 0.83832276,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86059737,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.715606927871704
    },
    {
      "auxiliary_loss_clip": 0.0123612,
      "auxiliary_loss_mlp": 0.01044821,
      "balance_loss_clip": 1.06804597,
      "balance_loss_mlp": 1.03269148,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.7431649035507704,
      "language_loss": 0.68173647,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70454586,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.7676851749420166
    },
    {
      "auxiliary_loss_clip": 0.0121238,
      "auxiliary_loss_mlp": 0.01043512,
      "balance_loss_clip": 1.06094539,
      "balance_loss_mlp": 1.03165674,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 2.0098668058474636,
      "language_loss": 0.74446929,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76702821,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.66540789604187
    },
    {
      "auxiliary_loss_clip": 0.01205898,
      "auxiliary_loss_mlp": 0.01042851,
      "balance_loss_clip": 1.05817842,
      "balance_loss_mlp": 1.03095388,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.8488883149769486,
      "language_loss": 0.79482913,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81731665,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.681696653366089
    },
    {
      "auxiliary_loss_clip": 0.01226192,
      "auxiliary_loss_mlp": 0.01050159,
      "balance_loss_clip": 1.0655545,
      "balance_loss_mlp": 1.03959072,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 2.1195379600974014,
      "language_loss": 0.78831553,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81107897,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.6513607501983643
    },
    {
      "auxiliary_loss_clip": 0.01187794,
      "auxiliary_loss_mlp": 0.01037934,
      "balance_loss_clip": 1.05697596,
      "balance_loss_mlp": 1.02623963,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.3292337505083234,
      "language_loss": 0.84901547,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.8712728,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.715359687805176
    },
    {
      "auxiliary_loss_clip": 0.01125401,
      "auxiliary_loss_mlp": 0.01016207,
      "balance_loss_clip": 1.03304863,
      "balance_loss_mlp": 1.01319075,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.850900353523007,
      "language_loss": 0.61609602,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63751209,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.157888650894165
    },
    {
      "auxiliary_loss_clip": 0.01123972,
      "auxiliary_loss_mlp": 0.01011496,
      "balance_loss_clip": 1.03236139,
      "balance_loss_mlp": 1.00837314,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8850230340786626,
      "language_loss": 0.62480569,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64616036,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.076507806777954
    },
    {
      "auxiliary_loss_clip": 0.01205582,
      "auxiliary_loss_mlp": 0.01040494,
      "balance_loss_clip": 1.05651593,
      "balance_loss_mlp": 1.02797711,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.0872184839836065,
      "language_loss": 0.77629638,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.7987572,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.7736904621124268
    },
    {
      "auxiliary_loss_clip": 0.01210419,
      "auxiliary_loss_mlp": 0.00906872,
      "balance_loss_clip": 1.06685376,
      "balance_loss_mlp": 1.00221491,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 2.2296544730418684,
      "language_loss": 0.78809249,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.80926538,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.7178540229797363
    },
    {
      "auxiliary_loss_clip": 0.01211724,
      "auxiliary_loss_mlp": 0.01046427,
      "balance_loss_clip": 1.06047153,
      "balance_loss_mlp": 1.03461933,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.404260670933899,
      "language_loss": 0.77312994,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79571146,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.704904079437256
    },
    {
      "auxiliary_loss_clip": 0.01213762,
      "auxiliary_loss_mlp": 0.01042537,
      "balance_loss_clip": 1.06634569,
      "balance_loss_mlp": 1.02973437,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.3453089992811758,
      "language_loss": 0.77951026,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80207324,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.671118974685669
    },
    {
      "auxiliary_loss_clip": 0.01204644,
      "auxiliary_loss_mlp": 0.01044454,
      "balance_loss_clip": 1.06590033,
      "balance_loss_mlp": 1.03371918,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 2.1115277630429845,
      "language_loss": 0.91506797,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93755889,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.674408197402954
    },
    {
      "auxiliary_loss_clip": 0.01223966,
      "auxiliary_loss_mlp": 0.01042666,
      "balance_loss_clip": 1.06352997,
      "balance_loss_mlp": 1.03126383,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 2.1045370397578487,
      "language_loss": 0.78605866,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80872494,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.699009895324707
    },
    {
      "auxiliary_loss_clip": 0.01233769,
      "auxiliary_loss_mlp": 0.01037921,
      "balance_loss_clip": 1.06552434,
      "balance_loss_mlp": 1.02574956,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.325052030884909,
      "language_loss": 0.79018682,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.8129037,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.6406431198120117
    },
    {
      "auxiliary_loss_clip": 0.01218878,
      "auxiliary_loss_mlp": 0.01036146,
      "balance_loss_clip": 1.06328249,
      "balance_loss_mlp": 1.02339077,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 2.079471165986303,
      "language_loss": 0.83401811,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85656834,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.617518663406372
    },
    {
      "auxiliary_loss_clip": 0.01228865,
      "auxiliary_loss_mlp": 0.01040869,
      "balance_loss_clip": 1.06389809,
      "balance_loss_mlp": 1.02863216,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 1.9494729688683055,
      "language_loss": 0.86242533,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88512278,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.6734042167663574
    },
    {
      "auxiliary_loss_clip": 0.01218513,
      "auxiliary_loss_mlp": 0.01046007,
      "balance_loss_clip": 1.06352699,
      "balance_loss_mlp": 1.03466403,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.976609591461128,
      "language_loss": 0.77858186,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80122697,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.6091148853302
    },
    {
      "auxiliary_loss_clip": 0.01232696,
      "auxiliary_loss_mlp": 0.01041332,
      "balance_loss_clip": 1.06560636,
      "balance_loss_mlp": 1.02964377,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.4096174631437584,
      "language_loss": 0.68100297,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70374322,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 3.492788314819336
    },
    {
      "auxiliary_loss_clip": 0.0122765,
      "auxiliary_loss_mlp": 0.01041149,
      "balance_loss_clip": 1.0650003,
      "balance_loss_mlp": 1.02907896,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 2.2924254355000246,
      "language_loss": 0.78434604,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80703402,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 3.7419540882110596
    },
    {
      "auxiliary_loss_clip": 0.01105049,
      "auxiliary_loss_mlp": 0.01011229,
      "balance_loss_clip": 1.02763605,
      "balance_loss_mlp": 1.0077486,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9273654603382866,
      "language_loss": 0.65715927,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67832208,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.2789371013641357
    },
    {
      "auxiliary_loss_clip": 0.01177036,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.05713022,
      "balance_loss_mlp": 1.02027845,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 2.3215558378042793,
      "language_loss": 0.83629,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85837078,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 3.770618200302124
    },
    {
      "auxiliary_loss_clip": 0.01208939,
      "auxiliary_loss_mlp": 0.01052793,
      "balance_loss_clip": 1.06058693,
      "balance_loss_mlp": 1.04035974,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.7625368124116507,
      "language_loss": 0.79446858,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81708586,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 3.5446596145629883
    },
    {
      "auxiliary_loss_clip": 0.01220731,
      "auxiliary_loss_mlp": 0.01037639,
      "balance_loss_clip": 1.06031942,
      "balance_loss_mlp": 1.02569437,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.026671643781126,
      "language_loss": 0.79483837,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81742203,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.6332859992980957
    },
    {
      "auxiliary_loss_clip": 0.0119959,
      "auxiliary_loss_mlp": 0.01039572,
      "balance_loss_clip": 1.05883157,
      "balance_loss_mlp": 1.02763319,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 1.9818123705837247,
      "language_loss": 0.74855292,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.77094448,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.760153293609619
    },
    {
      "auxiliary_loss_clip": 0.01215965,
      "auxiliary_loss_mlp": 0.01032591,
      "balance_loss_clip": 1.06440425,
      "balance_loss_mlp": 1.02043724,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 4.475617849561032,
      "language_loss": 0.80801177,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.83049738,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.6900436878204346
    },
    {
      "auxiliary_loss_clip": 0.01222724,
      "auxiliary_loss_mlp": 0.01040443,
      "balance_loss_clip": 1.06285357,
      "balance_loss_mlp": 1.02879643,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 2.5478416333581926,
      "language_loss": 0.76925141,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.79188311,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.5988311767578125
    },
    {
      "auxiliary_loss_clip": 0.01175499,
      "auxiliary_loss_mlp": 0.01037693,
      "balance_loss_clip": 1.05221987,
      "balance_loss_mlp": 1.02531266,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.3642622636221367,
      "language_loss": 0.7942996,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81643152,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.7060699462890625
    },
    {
      "auxiliary_loss_clip": 0.01239026,
      "auxiliary_loss_mlp": 0.01047111,
      "balance_loss_clip": 1.07193851,
      "balance_loss_mlp": 1.03514206,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 2.8446814196563808,
      "language_loss": 0.68030578,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70316714,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.613571882247925
    },
    {
      "auxiliary_loss_clip": 0.01224096,
      "auxiliary_loss_mlp": 0.00906228,
      "balance_loss_clip": 1.06619346,
      "balance_loss_mlp": 1.00208747,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 2.181944754427109,
      "language_loss": 0.82270527,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84400845,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.7017862796783447
    },
    {
      "auxiliary_loss_clip": 0.01171822,
      "auxiliary_loss_mlp": 0.01038433,
      "balance_loss_clip": 1.05185461,
      "balance_loss_mlp": 1.02590466,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 2.1018721033187475,
      "language_loss": 0.7783559,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80045843,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.7813973426818848
    },
    {
      "auxiliary_loss_clip": 0.01179836,
      "auxiliary_loss_mlp": 0.01048377,
      "balance_loss_clip": 1.05521834,
      "balance_loss_mlp": 1.03634906,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 2.724830092957805,
      "language_loss": 0.87915349,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90143561,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.7383017539978027
    },
    {
      "auxiliary_loss_clip": 0.01175355,
      "auxiliary_loss_mlp": 0.01043763,
      "balance_loss_clip": 1.05767095,
      "balance_loss_mlp": 1.03236032,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 4.477322207323649,
      "language_loss": 0.81620121,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83839238,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.756368398666382
    },
    {
      "auxiliary_loss_clip": 0.01219653,
      "auxiliary_loss_mlp": 0.01043443,
      "balance_loss_clip": 1.06185102,
      "balance_loss_mlp": 1.0307591,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.095473162104868,
      "language_loss": 0.77733362,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.79996455,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.6244304180145264
    },
    {
      "auxiliary_loss_clip": 0.01203554,
      "auxiliary_loss_mlp": 0.01049354,
      "balance_loss_clip": 1.06124783,
      "balance_loss_mlp": 1.0376054,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.853656269717731,
      "language_loss": 0.84522474,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86775386,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.627324104309082
    },
    {
      "auxiliary_loss_clip": 0.01196337,
      "auxiliary_loss_mlp": 0.01038068,
      "balance_loss_clip": 1.05516994,
      "balance_loss_mlp": 1.02711892,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 1.8154127513408251,
      "language_loss": 0.60228848,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62463254,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.897534132003784
    },
    {
      "auxiliary_loss_clip": 0.01221323,
      "auxiliary_loss_mlp": 0.01038994,
      "balance_loss_clip": 1.06154811,
      "balance_loss_mlp": 1.02758014,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.540523176180377,
      "language_loss": 0.80335426,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.82595742,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.6405344009399414
    },
    {
      "auxiliary_loss_clip": 0.01208947,
      "auxiliary_loss_mlp": 0.00905808,
      "balance_loss_clip": 1.06334078,
      "balance_loss_mlp": 1.00183141,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.0419264400543953,
      "language_loss": 0.86368775,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.8848353,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.657353639602661
    },
    {
      "auxiliary_loss_clip": 0.01200646,
      "auxiliary_loss_mlp": 0.01047078,
      "balance_loss_clip": 1.06578755,
      "balance_loss_mlp": 1.03548527,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.3571600529183057,
      "language_loss": 0.84368169,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86615902,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.789307117462158
    },
    {
      "auxiliary_loss_clip": 0.01199961,
      "auxiliary_loss_mlp": 0.01041993,
      "balance_loss_clip": 1.05904627,
      "balance_loss_mlp": 1.02985811,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.6728900943642637,
      "language_loss": 0.61043417,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63285375,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.7128541469573975
    },
    {
      "auxiliary_loss_clip": 0.01207142,
      "auxiliary_loss_mlp": 0.0103972,
      "balance_loss_clip": 1.06236863,
      "balance_loss_mlp": 1.02882433,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.9603671874893098,
      "language_loss": 0.77879828,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.80126691,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.7469162940979004
    },
    {
      "auxiliary_loss_clip": 0.01235721,
      "auxiliary_loss_mlp": 0.01041537,
      "balance_loss_clip": 1.06896508,
      "balance_loss_mlp": 1.03003311,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 2.876055951707175,
      "language_loss": 0.75972414,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78249669,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 2.70406436920166
    },
    {
      "auxiliary_loss_clip": 0.01230641,
      "auxiliary_loss_mlp": 0.01028697,
      "balance_loss_clip": 1.06516552,
      "balance_loss_mlp": 1.01808774,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 1.9562468237300221,
      "language_loss": 0.80277693,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82537037,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.6543664932250977
    },
    {
      "auxiliary_loss_clip": 0.01182302,
      "auxiliary_loss_mlp": 0.01036717,
      "balance_loss_clip": 1.05634856,
      "balance_loss_mlp": 1.02577329,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.9485398932602225,
      "language_loss": 0.77890891,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.80109906,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.8124382495880127
    },
    {
      "auxiliary_loss_clip": 0.01199868,
      "auxiliary_loss_mlp": 0.01039775,
      "balance_loss_clip": 1.06367505,
      "balance_loss_mlp": 1.02777684,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 1.8491241786787957,
      "language_loss": 0.78918099,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.81157738,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.8869407176971436
    },
    {
      "auxiliary_loss_clip": 0.01228385,
      "auxiliary_loss_mlp": 0.01046141,
      "balance_loss_clip": 1.06490719,
      "balance_loss_mlp": 1.03544199,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.123445002096804,
      "language_loss": 0.82017565,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.84292078,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 3.5206174850463867
    },
    {
      "auxiliary_loss_clip": 0.01200769,
      "auxiliary_loss_mlp": 0.01036095,
      "balance_loss_clip": 1.06230593,
      "balance_loss_mlp": 1.02518177,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 2.391013171650295,
      "language_loss": 0.7239027,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74627131,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 3.6463866233825684
    },
    {
      "auxiliary_loss_clip": 0.01222565,
      "auxiliary_loss_mlp": 0.01042763,
      "balance_loss_clip": 1.06549263,
      "balance_loss_mlp": 1.03113401,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 3.690846063004018,
      "language_loss": 0.82235879,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84501207,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.594860792160034
    },
    {
      "auxiliary_loss_clip": 0.01207607,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.06356096,
      "balance_loss_mlp": 1.02450109,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 2.469843390982231,
      "language_loss": 0.72757286,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74999619,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 4.501689672470093
    },
    {
      "auxiliary_loss_clip": 0.01206314,
      "auxiliary_loss_mlp": 0.01039066,
      "balance_loss_clip": 1.05892527,
      "balance_loss_mlp": 1.02781308,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.3479886527562988,
      "language_loss": 0.82662141,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.84907526,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.65696382522583
    },
    {
      "auxiliary_loss_clip": 0.01217446,
      "auxiliary_loss_mlp": 0.01040934,
      "balance_loss_clip": 1.06180322,
      "balance_loss_mlp": 1.02981186,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 3.03388260287075,
      "language_loss": 0.83429658,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.85688037,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.7284457683563232
    },
    {
      "auxiliary_loss_clip": 0.01214428,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 1.05902624,
      "balance_loss_mlp": 1.02512217,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.7018450992343928,
      "language_loss": 0.8707512,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89324909,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.681292772293091
    },
    {
      "auxiliary_loss_clip": 0.0121202,
      "auxiliary_loss_mlp": 0.01040704,
      "balance_loss_clip": 1.0635134,
      "balance_loss_mlp": 1.02992129,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 1.7336012915642984,
      "language_loss": 0.86725795,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88978517,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.6940128803253174
    },
    {
      "auxiliary_loss_clip": 0.01189835,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.05831909,
      "balance_loss_mlp": 1.020262,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.545428450437157,
      "language_loss": 0.8456229,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86782575,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.6571719646453857
    },
    {
      "auxiliary_loss_clip": 0.01119776,
      "auxiliary_loss_mlp": 0.01009883,
      "balance_loss_clip": 1.03026271,
      "balance_loss_mlp": 1.0068078,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8558086203350711,
      "language_loss": 0.57931393,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60061049,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.187797784805298
    },
    {
      "auxiliary_loss_clip": 0.0120066,
      "auxiliary_loss_mlp": 0.01041353,
      "balance_loss_clip": 1.05660224,
      "balance_loss_mlp": 1.0305469,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.362744725050804,
      "language_loss": 0.86261821,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88503838,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.68233060836792
    },
    {
      "auxiliary_loss_clip": 0.01192729,
      "auxiliary_loss_mlp": 0.01043229,
      "balance_loss_clip": 1.05930209,
      "balance_loss_mlp": 1.03188002,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.530045770346136,
      "language_loss": 0.78794104,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81030059,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.688162326812744
    },
    {
      "auxiliary_loss_clip": 0.01090937,
      "auxiliary_loss_mlp": 0.00893815,
      "balance_loss_clip": 1.03032351,
      "balance_loss_mlp": 1.0004549,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8624167805800017,
      "language_loss": 0.57907093,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.59891844,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.266284465789795
    },
    {
      "auxiliary_loss_clip": 0.01221144,
      "auxiliary_loss_mlp": 0.01038875,
      "balance_loss_clip": 1.06703043,
      "balance_loss_mlp": 1.02831936,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 3.553083912026448,
      "language_loss": 0.80484605,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.82744616,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.624035120010376
    },
    {
      "auxiliary_loss_clip": 0.01219062,
      "auxiliary_loss_mlp": 0.0103951,
      "balance_loss_clip": 1.06375778,
      "balance_loss_mlp": 1.02834594,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 4.063665537404491,
      "language_loss": 0.79258978,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81517553,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.6839113235473633
    },
    {
      "auxiliary_loss_clip": 0.0120011,
      "auxiliary_loss_mlp": 0.01038922,
      "balance_loss_clip": 1.05776906,
      "balance_loss_mlp": 1.02795434,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.395140515471348,
      "language_loss": 0.76020646,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78259683,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.7424585819244385
    },
    {
      "auxiliary_loss_clip": 0.01228107,
      "auxiliary_loss_mlp": 0.01038586,
      "balance_loss_clip": 1.06544125,
      "balance_loss_mlp": 1.0269568,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 2.1267668754112408,
      "language_loss": 0.74963391,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77230084,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.6049370765686035
    },
    {
      "auxiliary_loss_clip": 0.01217877,
      "auxiliary_loss_mlp": 0.01033579,
      "balance_loss_clip": 1.06432092,
      "balance_loss_mlp": 1.02306437,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.2658049413431596,
      "language_loss": 0.69254094,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71505547,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.7048370838165283
    },
    {
      "auxiliary_loss_clip": 0.01174533,
      "auxiliary_loss_mlp": 0.01036645,
      "balance_loss_clip": 1.05694485,
      "balance_loss_mlp": 1.02650666,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.3460168335211846,
      "language_loss": 0.88705707,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.90916884,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.777367115020752
    },
    {
      "auxiliary_loss_clip": 0.0122362,
      "auxiliary_loss_mlp": 0.01041336,
      "balance_loss_clip": 1.06470942,
      "balance_loss_mlp": 1.03042829,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.2231207587629203,
      "language_loss": 0.72533548,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.74798501,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.744921922683716
    },
    {
      "auxiliary_loss_clip": 0.01212845,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.06028748,
      "balance_loss_mlp": 1.0211755,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 3.8590723629891617,
      "language_loss": 0.77910316,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.8015604,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.6757657527923584
    },
    {
      "auxiliary_loss_clip": 0.01214872,
      "auxiliary_loss_mlp": 0.01034383,
      "balance_loss_clip": 1.06240344,
      "balance_loss_mlp": 1.02343321,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.5440308029681074,
      "language_loss": 0.88178414,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90427673,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.694251775741577
    },
    {
      "auxiliary_loss_clip": 0.01215878,
      "auxiliary_loss_mlp": 0.0103289,
      "balance_loss_clip": 1.06451726,
      "balance_loss_mlp": 1.02170825,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.5781658115448356,
      "language_loss": 0.8092373,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.831725,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.6130170822143555
    },
    {
      "auxiliary_loss_clip": 0.01184191,
      "auxiliary_loss_mlp": 0.01035819,
      "balance_loss_clip": 1.06037855,
      "balance_loss_mlp": 1.02464938,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.9352512609933088,
      "language_loss": 0.75755203,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.77975214,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.7630703449249268
    },
    {
      "auxiliary_loss_clip": 0.01216903,
      "auxiliary_loss_mlp": 0.01033239,
      "balance_loss_clip": 1.06396484,
      "balance_loss_mlp": 1.02215278,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 3.6094776417451144,
      "language_loss": 0.78012794,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.80262935,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.706472873687744
    },
    {
      "auxiliary_loss_clip": 0.01214577,
      "auxiliary_loss_mlp": 0.0104096,
      "balance_loss_clip": 1.06237173,
      "balance_loss_mlp": 1.02990305,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 3.2033717865639644,
      "language_loss": 0.82641846,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.84897381,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.64200496673584
    },
    {
      "auxiliary_loss_clip": 0.01228598,
      "auxiliary_loss_mlp": 0.01032242,
      "balance_loss_clip": 1.06836963,
      "balance_loss_mlp": 1.02226973,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.3187543581004197,
      "language_loss": 0.85334527,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87595367,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 3.4545319080352783
    },
    {
      "auxiliary_loss_clip": 0.01230617,
      "auxiliary_loss_mlp": 0.00904789,
      "balance_loss_clip": 1.06833005,
      "balance_loss_mlp": 1.00177252,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 3.199388150174955,
      "language_loss": 0.69786477,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.71921885,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 3.533186435699463
    },
    {
      "auxiliary_loss_clip": 0.01207902,
      "auxiliary_loss_mlp": 0.01034149,
      "balance_loss_clip": 1.06255698,
      "balance_loss_mlp": 1.02390265,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.5736620813396744,
      "language_loss": 0.78926122,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81168169,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.6698555946350098
    },
    {
      "auxiliary_loss_clip": 0.01181531,
      "auxiliary_loss_mlp": 0.0103663,
      "balance_loss_clip": 1.05576158,
      "balance_loss_mlp": 1.02482224,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.2226096693712813,
      "language_loss": 0.94865191,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.97083354,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 3.6830661296844482
    },
    {
      "auxiliary_loss_clip": 0.01230221,
      "auxiliary_loss_mlp": 0.01034821,
      "balance_loss_clip": 1.0686475,
      "balance_loss_mlp": 1.02452683,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.987787418631544,
      "language_loss": 0.85139751,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87404794,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 3.7081072330474854
    },
    {
      "auxiliary_loss_clip": 0.01218715,
      "auxiliary_loss_mlp": 0.01038467,
      "balance_loss_clip": 1.0655508,
      "balance_loss_mlp": 1.02758908,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.3769901041181236,
      "language_loss": 0.866036,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.8886078,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.619924306869507
    },
    {
      "auxiliary_loss_clip": 0.01209718,
      "auxiliary_loss_mlp": 0.01036398,
      "balance_loss_clip": 1.06761074,
      "balance_loss_mlp": 1.02530599,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 1.991406322823994,
      "language_loss": 0.80223441,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82469559,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.6396026611328125
    },
    {
      "auxiliary_loss_clip": 0.01202595,
      "auxiliary_loss_mlp": 0.01035066,
      "balance_loss_clip": 1.0639807,
      "balance_loss_mlp": 1.02439094,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 2.0023401232222295,
      "language_loss": 0.77963793,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80201447,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.7788329124450684
    },
    {
      "auxiliary_loss_clip": 0.01194053,
      "auxiliary_loss_mlp": 0.0103561,
      "balance_loss_clip": 1.05995512,
      "balance_loss_mlp": 1.02503014,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 4.3319750371457655,
      "language_loss": 0.93816507,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.96046162,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.605184555053711
    },
    {
      "auxiliary_loss_clip": 0.01197631,
      "auxiliary_loss_mlp": 0.01037437,
      "balance_loss_clip": 1.05802715,
      "balance_loss_mlp": 1.0264163,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 1.9225602685772503,
      "language_loss": 0.75505471,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77740544,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.7271981239318848
    },
    {
      "auxiliary_loss_clip": 0.01216642,
      "auxiliary_loss_mlp": 0.01036941,
      "balance_loss_clip": 1.06141281,
      "balance_loss_mlp": 1.02680182,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 2.418327119878985,
      "language_loss": 0.85631251,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.87884831,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.6494059562683105
    },
    {
      "auxiliary_loss_clip": 0.01180453,
      "auxiliary_loss_mlp": 0.01044499,
      "balance_loss_clip": 1.05902851,
      "balance_loss_mlp": 1.03335261,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 1.9400041802752226,
      "language_loss": 0.87856823,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.90081775,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.7230241298675537
    },
    {
      "auxiliary_loss_clip": 0.0120845,
      "auxiliary_loss_mlp": 0.01033458,
      "balance_loss_clip": 1.06302428,
      "balance_loss_mlp": 1.02151299,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 2.1646821443913846,
      "language_loss": 0.73622406,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75864309,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.6781957149505615
    },
    {
      "auxiliary_loss_clip": 0.01214707,
      "auxiliary_loss_mlp": 0.01038109,
      "balance_loss_clip": 1.06230533,
      "balance_loss_mlp": 1.02735674,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 1.9272041225816896,
      "language_loss": 0.85150594,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87403411,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.6268835067749023
    },
    {
      "auxiliary_loss_clip": 0.01117514,
      "auxiliary_loss_mlp": 0.01005346,
      "balance_loss_clip": 1.03490448,
      "balance_loss_mlp": 1.00277126,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.8235443194550506,
      "language_loss": 0.57351702,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59474564,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.236750364303589
    },
    {
      "auxiliary_loss_clip": 0.01168349,
      "auxiliary_loss_mlp": 0.01037809,
      "balance_loss_clip": 1.0581187,
      "balance_loss_mlp": 1.02693081,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 10.843400876726271,
      "language_loss": 0.79104388,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.81310546,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.754532814025879
    },
    {
      "auxiliary_loss_clip": 0.01203645,
      "auxiliary_loss_mlp": 0.01038671,
      "balance_loss_clip": 1.06146467,
      "balance_loss_mlp": 1.02754903,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 3.493401860076265,
      "language_loss": 0.77103412,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.79345727,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.628364086151123
    },
    {
      "auxiliary_loss_clip": 0.01162175,
      "auxiliary_loss_mlp": 0.01036717,
      "balance_loss_clip": 1.05630183,
      "balance_loss_mlp": 1.02610755,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.9763382303010313,
      "language_loss": 0.86445022,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88643909,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.7271888256073
    },
    {
      "auxiliary_loss_clip": 0.01092001,
      "auxiliary_loss_mlp": 0.01000337,
      "balance_loss_clip": 1.02896953,
      "balance_loss_mlp": 0.99766713,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8463102645141691,
      "language_loss": 0.61496359,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63588697,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.2387940883636475
    },
    {
      "auxiliary_loss_clip": 0.01206302,
      "auxiliary_loss_mlp": 0.01042262,
      "balance_loss_clip": 1.06454289,
      "balance_loss_mlp": 1.03141427,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.304697297248706,
      "language_loss": 0.86129081,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88377643,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.6044371128082275
    },
    {
      "auxiliary_loss_clip": 0.01187477,
      "auxiliary_loss_mlp": 0.01037774,
      "balance_loss_clip": 1.05523944,
      "balance_loss_mlp": 1.02714658,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.3011377798257806,
      "language_loss": 0.8369807,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.8592332,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.6994845867156982
    },
    {
      "auxiliary_loss_clip": 0.01215413,
      "auxiliary_loss_mlp": 0.01041617,
      "balance_loss_clip": 1.06503892,
      "balance_loss_mlp": 1.03101981,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 1.923789840822468,
      "language_loss": 0.82317793,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84574819,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 2.6798043251037598
    },
    {
      "auxiliary_loss_clip": 0.01213189,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.06761241,
      "balance_loss_mlp": 1.0190661,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 2.2300115894465224,
      "language_loss": 0.74489617,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76732552,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.719822645187378
    },
    {
      "auxiliary_loss_clip": 0.01111529,
      "auxiliary_loss_mlp": 0.01008092,
      "balance_loss_clip": 1.02972817,
      "balance_loss_mlp": 1.0054096,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.809449596900273,
      "language_loss": 0.60426033,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62545657,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.195648670196533
    },
    {
      "auxiliary_loss_clip": 0.01196358,
      "auxiliary_loss_mlp": 0.01034499,
      "balance_loss_clip": 1.05656838,
      "balance_loss_mlp": 1.02424693,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.7941341754143942,
      "language_loss": 0.74340224,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76571071,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.811506986618042
    },
    {
      "auxiliary_loss_clip": 0.01209401,
      "auxiliary_loss_mlp": 0.01036866,
      "balance_loss_clip": 1.06533217,
      "balance_loss_mlp": 1.0262208,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 2.0554380253771094,
      "language_loss": 0.72396541,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74642813,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.7934491634368896
    },
    {
      "auxiliary_loss_clip": 0.01218939,
      "auxiliary_loss_mlp": 0.01037163,
      "balance_loss_clip": 1.06236565,
      "balance_loss_mlp": 1.02607024,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 2.2198460991201285,
      "language_loss": 0.78232348,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80488455,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.643869638442993
    },
    {
      "auxiliary_loss_clip": 0.01199889,
      "auxiliary_loss_mlp": 0.01032521,
      "balance_loss_clip": 1.06034946,
      "balance_loss_mlp": 1.02229309,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.542658165794511,
      "language_loss": 0.73933625,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76166034,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 3.707160234451294
    },
    {
      "auxiliary_loss_clip": 0.01101174,
      "auxiliary_loss_mlp": 0.01006173,
      "balance_loss_clip": 1.02575517,
      "balance_loss_mlp": 1.0034436,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8194474902865193,
      "language_loss": 0.60435849,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62543195,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.971405506134033
    },
    {
      "auxiliary_loss_clip": 0.01199742,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.05923522,
      "balance_loss_mlp": 1.02965474,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.7728267445933446,
      "language_loss": 0.76389945,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78629875,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 3.4959254264831543
    },
    {
      "auxiliary_loss_clip": 0.0121115,
      "auxiliary_loss_mlp": 0.01043533,
      "balance_loss_clip": 1.06102872,
      "balance_loss_mlp": 1.03171384,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 3.4472316879901603,
      "language_loss": 0.86515224,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.88769913,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 3.3813464641571045
    },
    {
      "auxiliary_loss_clip": 0.01202383,
      "auxiliary_loss_mlp": 0.01036565,
      "balance_loss_clip": 1.05944228,
      "balance_loss_mlp": 1.02608061,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.6857986583124718,
      "language_loss": 0.79413009,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81651962,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.699281930923462
    },
    {
      "auxiliary_loss_clip": 0.01216928,
      "auxiliary_loss_mlp": 0.01046375,
      "balance_loss_clip": 1.06698227,
      "balance_loss_mlp": 1.03429937,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 2.0665135669725534,
      "language_loss": 0.73139322,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75402617,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.5923781394958496
    },
    {
      "auxiliary_loss_clip": 0.01205702,
      "auxiliary_loss_mlp": 0.01038962,
      "balance_loss_clip": 1.06015182,
      "balance_loss_mlp": 1.02778006,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 2.294553311379689,
      "language_loss": 0.86872435,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89117098,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.673982858657837
    },
    {
      "auxiliary_loss_clip": 0.01207627,
      "auxiliary_loss_mlp": 0.01035878,
      "balance_loss_clip": 1.0610218,
      "balance_loss_mlp": 1.0248754,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.6338777397718232,
      "language_loss": 0.73841667,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76085174,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.821889638900757
    },
    {
      "auxiliary_loss_clip": 0.0119958,
      "auxiliary_loss_mlp": 0.01039111,
      "balance_loss_clip": 1.06061184,
      "balance_loss_mlp": 1.02739859,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 2.8531083533892683,
      "language_loss": 0.79157197,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.81395888,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.631042957305908
    },
    {
      "auxiliary_loss_clip": 0.01207927,
      "auxiliary_loss_mlp": 0.01044294,
      "balance_loss_clip": 1.06071877,
      "balance_loss_mlp": 1.03363037,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.012215842619477,
      "language_loss": 0.91625875,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93878096,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.6586222648620605
    },
    {
      "auxiliary_loss_clip": 0.01186485,
      "auxiliary_loss_mlp": 0.01043775,
      "balance_loss_clip": 1.05291283,
      "balance_loss_mlp": 1.03239667,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.8080259897814803,
      "language_loss": 0.90259159,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92489421,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.797581911087036
    },
    {
      "auxiliary_loss_clip": 0.01204677,
      "auxiliary_loss_mlp": 0.01041083,
      "balance_loss_clip": 1.06324589,
      "balance_loss_mlp": 1.03021669,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.5519742774252103,
      "language_loss": 0.79823422,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.82069182,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.9019181728363037
    },
    {
      "auxiliary_loss_clip": 0.01207714,
      "auxiliary_loss_mlp": 0.01039142,
      "balance_loss_clip": 1.06060719,
      "balance_loss_mlp": 1.02854478,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.82200990295143,
      "language_loss": 0.92760003,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.95006859,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.624957799911499
    },
    {
      "auxiliary_loss_clip": 0.01220156,
      "auxiliary_loss_mlp": 0.01040222,
      "balance_loss_clip": 1.06123519,
      "balance_loss_mlp": 1.02906966,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 3.057534968208241,
      "language_loss": 0.78520155,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.8078053,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.624298334121704
    },
    {
      "auxiliary_loss_clip": 0.01197432,
      "auxiliary_loss_mlp": 0.01042137,
      "balance_loss_clip": 1.06074345,
      "balance_loss_mlp": 1.03015637,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.198305323466287,
      "language_loss": 0.8170377,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83943337,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.691500425338745
    },
    {
      "auxiliary_loss_clip": 0.01208471,
      "auxiliary_loss_mlp": 0.01036298,
      "balance_loss_clip": 1.05954349,
      "balance_loss_mlp": 1.0237515,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.7999792835268265,
      "language_loss": 0.81676918,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83921683,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.6080753803253174
    },
    {
      "auxiliary_loss_clip": 0.01158793,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.04937565,
      "balance_loss_mlp": 1.02517617,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 2.0044203124902262,
      "language_loss": 0.84811163,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.87005967,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.70489501953125
    },
    {
      "auxiliary_loss_clip": 0.01201667,
      "auxiliary_loss_mlp": 0.01041188,
      "balance_loss_clip": 1.05604196,
      "balance_loss_mlp": 1.02983284,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.0180176120492206,
      "language_loss": 0.86076546,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.88319397,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.7026309967041016
    },
    {
      "auxiliary_loss_clip": 0.01116783,
      "auxiliary_loss_mlp": 0.0100233,
      "balance_loss_clip": 1.02793777,
      "balance_loss_mlp": 0.99994612,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7464914413036801,
      "language_loss": 0.61200798,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.6331991,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.2737438678741455
    },
    {
      "auxiliary_loss_clip": 0.01193134,
      "auxiliary_loss_mlp": 0.01038251,
      "balance_loss_clip": 1.05900049,
      "balance_loss_mlp": 1.0273428,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 2.0956469625367062,
      "language_loss": 0.75742233,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.77973616,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.7149198055267334
    },
    {
      "auxiliary_loss_clip": 0.01183784,
      "auxiliary_loss_mlp": 0.01035606,
      "balance_loss_clip": 1.05357122,
      "balance_loss_mlp": 1.0239656,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.2403540371335597,
      "language_loss": 0.74315739,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.7653513,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.7864644527435303
    },
    {
      "auxiliary_loss_clip": 0.01104346,
      "auxiliary_loss_mlp": 0.01004783,
      "balance_loss_clip": 1.0232892,
      "balance_loss_mlp": 1.00237489,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8389674774528442,
      "language_loss": 0.64487875,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66597003,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.0871291160583496
    },
    {
      "auxiliary_loss_clip": 0.01228139,
      "auxiliary_loss_mlp": 0.01034717,
      "balance_loss_clip": 1.06469095,
      "balance_loss_mlp": 1.02409506,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.1530004203906388,
      "language_loss": 0.82566106,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.84828961,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.7905213832855225
    },
    {
      "auxiliary_loss_clip": 0.0117991,
      "auxiliary_loss_mlp": 0.00904505,
      "balance_loss_clip": 1.05620956,
      "balance_loss_mlp": 1.00237036,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.212611435558149,
      "language_loss": 0.79886305,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.81970716,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.8115973472595215
    },
    {
      "auxiliary_loss_clip": 0.01209373,
      "auxiliary_loss_mlp": 0.0090536,
      "balance_loss_clip": 1.06320357,
      "balance_loss_mlp": 1.00238502,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 3.087165643661662,
      "language_loss": 0.77787447,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.79902172,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.7380318641662598
    },
    {
      "auxiliary_loss_clip": 0.01199397,
      "auxiliary_loss_mlp": 0.01033399,
      "balance_loss_clip": 1.06076717,
      "balance_loss_mlp": 1.02323079,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 1.9038960215797283,
      "language_loss": 0.76415539,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78648329,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.6563713550567627
    },
    {
      "auxiliary_loss_clip": 0.01218233,
      "auxiliary_loss_mlp": 0.01035682,
      "balance_loss_clip": 1.06159747,
      "balance_loss_mlp": 1.02478659,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.1677154215776726,
      "language_loss": 0.76507056,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.7876097,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.6537771224975586
    },
    {
      "auxiliary_loss_clip": 0.01162023,
      "auxiliary_loss_mlp": 0.01036047,
      "balance_loss_clip": 1.04700303,
      "balance_loss_mlp": 1.02523518,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.6657073587344196,
      "language_loss": 0.74113309,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76311374,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 4.63373875617981
    },
    {
      "auxiliary_loss_clip": 0.01166568,
      "auxiliary_loss_mlp": 0.00904589,
      "balance_loss_clip": 1.0532304,
      "balance_loss_mlp": 1.00238848,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.5384716095961428,
      "language_loss": 0.78322107,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.80393261,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.7856414318084717
    },
    {
      "auxiliary_loss_clip": 0.01106905,
      "auxiliary_loss_mlp": 0.01001942,
      "balance_loss_clip": 1.0234704,
      "balance_loss_mlp": 0.99958116,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9238446413542966,
      "language_loss": 0.57190931,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59299779,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 4.248589038848877
    },
    {
      "auxiliary_loss_clip": 0.01192083,
      "auxiliary_loss_mlp": 0.0090466,
      "balance_loss_clip": 1.05470872,
      "balance_loss_mlp": 1.00224161,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.1942715571752713,
      "language_loss": 0.78599721,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.80696464,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 3.5377743244171143
    },
    {
      "auxiliary_loss_clip": 0.01190647,
      "auxiliary_loss_mlp": 0.01038701,
      "balance_loss_clip": 1.06150961,
      "balance_loss_mlp": 1.028139,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 3.6523966211168184,
      "language_loss": 0.91768068,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.93997413,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.7020092010498047
    },
    {
      "auxiliary_loss_clip": 0.01217726,
      "auxiliary_loss_mlp": 0.01038182,
      "balance_loss_clip": 1.0642302,
      "balance_loss_mlp": 1.02760196,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 1.9347972196502414,
      "language_loss": 0.84473825,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86729723,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.7299399375915527
    },
    {
      "auxiliary_loss_clip": 0.01187676,
      "auxiliary_loss_mlp": 0.01043597,
      "balance_loss_clip": 1.05318165,
      "balance_loss_mlp": 1.03239119,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.4559765759619254,
      "language_loss": 0.83472204,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.8570348,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.6722350120544434
    },
    {
      "auxiliary_loss_clip": 0.01209362,
      "auxiliary_loss_mlp": 0.01040461,
      "balance_loss_clip": 1.05674744,
      "balance_loss_mlp": 1.02967215,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 1.903195066945752,
      "language_loss": 0.72775841,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.75025666,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.5937225818634033
    },
    {
      "auxiliary_loss_clip": 0.01215135,
      "auxiliary_loss_mlp": 0.01034999,
      "balance_loss_clip": 1.06327629,
      "balance_loss_mlp": 1.02401423,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 1.7731304267576644,
      "language_loss": 0.70321697,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72571832,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.6737184524536133
    },
    {
      "auxiliary_loss_clip": 0.01199682,
      "auxiliary_loss_mlp": 0.01034476,
      "balance_loss_clip": 1.05693316,
      "balance_loss_mlp": 1.02417672,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 3.852693867373048,
      "language_loss": 0.83497822,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85731983,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.6417198181152344
    },
    {
      "auxiliary_loss_clip": 0.01221903,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.06427717,
      "balance_loss_mlp": 1.02335167,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.8704953748578728,
      "language_loss": 0.80965877,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83222491,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.6634490489959717
    },
    {
      "auxiliary_loss_clip": 0.01159036,
      "auxiliary_loss_mlp": 0.01035267,
      "balance_loss_clip": 1.05036378,
      "balance_loss_mlp": 1.02393007,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 1.9569542826509454,
      "language_loss": 0.70334029,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72528338,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.8136019706726074
    },
    {
      "auxiliary_loss_clip": 0.01158961,
      "auxiliary_loss_mlp": 0.00905561,
      "balance_loss_clip": 1.04964662,
      "balance_loss_mlp": 1.00231874,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 1.9228055711875849,
      "language_loss": 0.75472575,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77537096,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.8758554458618164
    },
    {
      "auxiliary_loss_clip": 0.01213944,
      "auxiliary_loss_mlp": 0.0104354,
      "balance_loss_clip": 1.06377482,
      "balance_loss_mlp": 1.03210235,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 2.1886352963040387,
      "language_loss": 0.77592438,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79849923,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.6195268630981445
    },
    {
      "auxiliary_loss_clip": 0.01208627,
      "auxiliary_loss_mlp": 0.01034692,
      "balance_loss_clip": 1.05853868,
      "balance_loss_mlp": 1.02300346,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.374807313145503,
      "language_loss": 0.90717328,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.9296065,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.600024938583374
    },
    {
      "auxiliary_loss_clip": 0.01214814,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.05970883,
      "balance_loss_mlp": 1.01892376,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 2.4965109376141936,
      "language_loss": 0.8406738,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86311495,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.6385090351104736
    },
    {
      "auxiliary_loss_clip": 0.01226508,
      "auxiliary_loss_mlp": 0.01042979,
      "balance_loss_clip": 1.0644834,
      "balance_loss_mlp": 1.0324887,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.7363535715716654,
      "language_loss": 0.9163965,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.93909144,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.675020933151245
    },
    {
      "auxiliary_loss_clip": 0.01193166,
      "auxiliary_loss_mlp": 0.01043629,
      "balance_loss_clip": 1.05701363,
      "balance_loss_mlp": 1.03188133,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 1.819395459072261,
      "language_loss": 0.62858731,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.6509552,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 2.704332113265991
    },
    {
      "auxiliary_loss_clip": 0.01215234,
      "auxiliary_loss_mlp": 0.01030437,
      "balance_loss_clip": 1.05636454,
      "balance_loss_mlp": 1.01961243,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 3.191780577427632,
      "language_loss": 0.73904139,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.76149809,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.599726438522339
    },
    {
      "auxiliary_loss_clip": 0.0119517,
      "auxiliary_loss_mlp": 0.01037897,
      "balance_loss_clip": 1.05772805,
      "balance_loss_mlp": 1.02569032,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 1.9859618404882768,
      "language_loss": 0.75052619,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.77285683,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.7508044242858887
    },
    {
      "auxiliary_loss_clip": 0.0117591,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.05556202,
      "balance_loss_mlp": 1.02078378,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 2.409438128748836,
      "language_loss": 0.72802448,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.75009328,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.864060401916504
    },
    {
      "auxiliary_loss_clip": 0.01220168,
      "auxiliary_loss_mlp": 0.01046,
      "balance_loss_clip": 1.0624485,
      "balance_loss_mlp": 1.03463924,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.144114551335791,
      "language_loss": 0.78200299,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80466473,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.80118989944458
    },
    {
      "auxiliary_loss_clip": 0.01199793,
      "auxiliary_loss_mlp": 0.01035488,
      "balance_loss_clip": 1.05941439,
      "balance_loss_mlp": 1.02470517,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 2.0665030088669734,
      "language_loss": 0.78036612,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.80271888,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.671720504760742
    },
    {
      "auxiliary_loss_clip": 0.01201995,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.05842042,
      "balance_loss_mlp": 1.02476597,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.7931750189707976,
      "language_loss": 0.75815833,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.7805323,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.7333147525787354
    },
    {
      "auxiliary_loss_clip": 0.01205065,
      "auxiliary_loss_mlp": 0.01036357,
      "balance_loss_clip": 1.05902636,
      "balance_loss_mlp": 1.0263257,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.069330089494458,
      "language_loss": 0.72474957,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74716377,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.6193854808807373
    },
    {
      "auxiliary_loss_clip": 0.01205022,
      "auxiliary_loss_mlp": 0.01037394,
      "balance_loss_clip": 1.05949771,
      "balance_loss_mlp": 1.02605152,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 1.9669483156959902,
      "language_loss": 0.82310009,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84552425,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.661384105682373
    },
    {
      "auxiliary_loss_clip": 0.01186612,
      "auxiliary_loss_mlp": 0.01036502,
      "balance_loss_clip": 1.0573411,
      "balance_loss_mlp": 1.0251354,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.2975739871829686,
      "language_loss": 0.87675929,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89899039,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 3.657109498977661
    },
    {
      "auxiliary_loss_clip": 0.01170966,
      "auxiliary_loss_mlp": 0.01034336,
      "balance_loss_clip": 1.05110574,
      "balance_loss_mlp": 1.02397108,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.7788749857929105,
      "language_loss": 0.86632383,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88837689,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 3.624753475189209
    },
    {
      "auxiliary_loss_clip": 0.0122437,
      "auxiliary_loss_mlp": 0.01037753,
      "balance_loss_clip": 1.06446266,
      "balance_loss_mlp": 1.02767956,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 2.0177456966743,
      "language_loss": 0.82285643,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84547764,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.6125388145446777
    },
    {
      "auxiliary_loss_clip": 0.01187197,
      "auxiliary_loss_mlp": 0.01043991,
      "balance_loss_clip": 1.05710196,
      "balance_loss_mlp": 1.03226078,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 3.0054215688743318,
      "language_loss": 0.87035608,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.89266801,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 4.534734487533569
    },
    {
      "auxiliary_loss_clip": 0.01221683,
      "auxiliary_loss_mlp": 0.01034635,
      "balance_loss_clip": 1.06067562,
      "balance_loss_mlp": 1.0241034,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 2.1619242716789087,
      "language_loss": 0.76664758,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.7892108,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.835764169692993
    },
    {
      "auxiliary_loss_clip": 0.01191205,
      "auxiliary_loss_mlp": 0.00905576,
      "balance_loss_clip": 1.05823565,
      "balance_loss_mlp": 1.00209808,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.795854663327903,
      "language_loss": 0.78197217,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.80294001,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.711181402206421
    },
    {
      "auxiliary_loss_clip": 0.0109602,
      "auxiliary_loss_mlp": 0.00894766,
      "balance_loss_clip": 1.02526689,
      "balance_loss_mlp": 1.00076985,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7353615868647685,
      "language_loss": 0.60356522,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62347317,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.236345052719116
    },
    {
      "auxiliary_loss_clip": 0.01199731,
      "auxiliary_loss_mlp": 0.01035322,
      "balance_loss_clip": 1.05693865,
      "balance_loss_mlp": 1.02428913,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.8143341382131015,
      "language_loss": 0.79042691,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81277746,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.7089269161224365
    },
    {
      "auxiliary_loss_clip": 0.01198526,
      "auxiliary_loss_mlp": 0.01039718,
      "balance_loss_clip": 1.06076503,
      "balance_loss_mlp": 1.02853608,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.4309522122407206,
      "language_loss": 0.90954447,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.93192685,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.6754209995269775
    },
    {
      "auxiliary_loss_clip": 0.01189709,
      "auxiliary_loss_mlp": 0.0103695,
      "balance_loss_clip": 1.05216849,
      "balance_loss_mlp": 1.02569044,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 3.0838640249745732,
      "language_loss": 0.78098208,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.8032487,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.7380239963531494
    },
    {
      "auxiliary_loss_clip": 0.01228198,
      "auxiliary_loss_mlp": 0.01045699,
      "balance_loss_clip": 1.06474268,
      "balance_loss_mlp": 1.03389692,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 6.9066656054953715,
      "language_loss": 0.79666567,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.8194046,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.659235715866089
    },
    {
      "auxiliary_loss_clip": 0.01214806,
      "auxiliary_loss_mlp": 0.01036086,
      "balance_loss_clip": 1.06238115,
      "balance_loss_mlp": 1.02524376,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 2.277887646090724,
      "language_loss": 0.76590723,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.78841615,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.7347538471221924
    },
    {
      "auxiliary_loss_clip": 0.0118226,
      "auxiliary_loss_mlp": 0.00905372,
      "balance_loss_clip": 1.05625176,
      "balance_loss_mlp": 1.00205708,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.669080245732764,
      "language_loss": 0.85798037,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87885666,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.7817163467407227
    },
    {
      "auxiliary_loss_clip": 0.01221558,
      "auxiliary_loss_mlp": 0.0104041,
      "balance_loss_clip": 1.06093061,
      "balance_loss_mlp": 1.02987218,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.62939723555168,
      "language_loss": 0.81831634,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84093601,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.588376760482788
    },
    {
      "auxiliary_loss_clip": 0.01215511,
      "auxiliary_loss_mlp": 0.01037567,
      "balance_loss_clip": 1.05987382,
      "balance_loss_mlp": 1.0267427,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.6663080192892865,
      "language_loss": 0.79586792,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.81839877,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.647282838821411
    },
    {
      "auxiliary_loss_clip": 0.01195478,
      "auxiliary_loss_mlp": 0.01037404,
      "balance_loss_clip": 1.05374742,
      "balance_loss_mlp": 1.02704513,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 2.0287028707993477,
      "language_loss": 0.85144436,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87377316,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.760744094848633
    },
    {
      "auxiliary_loss_clip": 0.01195836,
      "auxiliary_loss_mlp": 0.01039793,
      "balance_loss_clip": 1.05923629,
      "balance_loss_mlp": 1.02914751,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.8871996763566758,
      "language_loss": 0.83657992,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85893619,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.71866774559021
    },
    {
      "auxiliary_loss_clip": 0.01219742,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.06272578,
      "balance_loss_mlp": 1.02509165,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.9667698609168847,
      "language_loss": 0.7145921,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73713779,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 2.510636806488037
    },
    {
      "auxiliary_loss_clip": 0.01199572,
      "auxiliary_loss_mlp": 0.010411,
      "balance_loss_clip": 1.06158495,
      "balance_loss_mlp": 1.02970994,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 1.9793452431566507,
      "language_loss": 0.83626866,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.85867536,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.698464870452881
    },
    {
      "auxiliary_loss_clip": 0.01202795,
      "auxiliary_loss_mlp": 0.01035293,
      "balance_loss_clip": 1.06036949,
      "balance_loss_mlp": 1.02421284,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.7073453623158732,
      "language_loss": 0.87029672,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89267766,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.6618897914886475
    },
    {
      "auxiliary_loss_clip": 0.01218172,
      "auxiliary_loss_mlp": 0.01033973,
      "balance_loss_clip": 1.06179726,
      "balance_loss_mlp": 1.02200472,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.0530400094259833,
      "language_loss": 0.71958482,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.74210632,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.5885579586029053
    },
    {
      "auxiliary_loss_clip": 0.01214128,
      "auxiliary_loss_mlp": 0.01043978,
      "balance_loss_clip": 1.05909681,
      "balance_loss_mlp": 1.03436995,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.6893437873614925,
      "language_loss": 0.80582166,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82840264,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.652946949005127
    },
    {
      "auxiliary_loss_clip": 0.01180839,
      "auxiliary_loss_mlp": 0.00905057,
      "balance_loss_clip": 1.05531538,
      "balance_loss_mlp": 1.00203884,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 1.9510127248024836,
      "language_loss": 0.80846167,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.82932067,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.699136734008789
    },
    {
      "auxiliary_loss_clip": 0.01212959,
      "auxiliary_loss_mlp": 0.00904663,
      "balance_loss_clip": 1.06120157,
      "balance_loss_mlp": 1.00195718,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 2.178648878689205,
      "language_loss": 0.84868282,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.8698591,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.633120059967041
    },
    {
      "auxiliary_loss_clip": 0.01201791,
      "auxiliary_loss_mlp": 0.01036209,
      "balance_loss_clip": 1.06080914,
      "balance_loss_mlp": 1.02561128,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 2.1444316279427182,
      "language_loss": 0.82967013,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.85205007,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.7213151454925537
    },
    {
      "auxiliary_loss_clip": 0.0118757,
      "auxiliary_loss_mlp": 0.00904976,
      "balance_loss_clip": 1.05781186,
      "balance_loss_mlp": 1.00183177,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 3.709449080324576,
      "language_loss": 0.72209972,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.74302518,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.7335562705993652
    },
    {
      "auxiliary_loss_clip": 0.01184394,
      "auxiliary_loss_mlp": 0.01035507,
      "balance_loss_clip": 1.05363727,
      "balance_loss_mlp": 1.02515388,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 2.6006458608366017,
      "language_loss": 0.83928132,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86148036,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.672380208969116
    },
    {
      "auxiliary_loss_clip": 0.01212601,
      "auxiliary_loss_mlp": 0.01035992,
      "balance_loss_clip": 1.06084013,
      "balance_loss_mlp": 1.02472711,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 4.05074539740672,
      "language_loss": 0.92718738,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.9496733,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 3.6014816761016846
    },
    {
      "auxiliary_loss_clip": 0.01194919,
      "auxiliary_loss_mlp": 0.0103313,
      "balance_loss_clip": 1.05878997,
      "balance_loss_mlp": 1.02297902,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.6829884607473367,
      "language_loss": 0.77066261,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79294312,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.7063982486724854
    },
    {
      "auxiliary_loss_clip": 0.0119347,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.0562973,
      "balance_loss_mlp": 1.02214313,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 3.162724101424137,
      "language_loss": 0.77193058,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.7941823,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.670539140701294
    },
    {
      "auxiliary_loss_clip": 0.01203455,
      "auxiliary_loss_mlp": 0.0103441,
      "balance_loss_clip": 1.06026173,
      "balance_loss_mlp": 1.02354431,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 8.214601409546189,
      "language_loss": 0.66976798,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.6921466,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 4.472872495651245
    },
    {
      "auxiliary_loss_clip": 0.01182519,
      "auxiliary_loss_mlp": 0.01037396,
      "balance_loss_clip": 1.0524019,
      "balance_loss_mlp": 1.02700639,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 2.975807657322413,
      "language_loss": 0.82781076,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.85000998,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.6431500911712646
    },
    {
      "auxiliary_loss_clip": 0.0122212,
      "auxiliary_loss_mlp": 0.01035381,
      "balance_loss_clip": 1.06013501,
      "balance_loss_mlp": 1.02362669,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 2.129931748377938,
      "language_loss": 0.82474983,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84732485,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.6811838150024414
    },
    {
      "auxiliary_loss_clip": 0.01204162,
      "auxiliary_loss_mlp": 0.0103682,
      "balance_loss_clip": 1.06083131,
      "balance_loss_mlp": 1.02578127,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 1.9395209066491956,
      "language_loss": 0.81314683,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83555663,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.76794695854187
    },
    {
      "auxiliary_loss_clip": 0.01213392,
      "auxiliary_loss_mlp": 0.01046875,
      "balance_loss_clip": 1.06295037,
      "balance_loss_mlp": 1.03553247,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 3.2425367209172387,
      "language_loss": 0.75991392,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.7825166,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.6420159339904785
    },
    {
      "auxiliary_loss_clip": 0.01205082,
      "auxiliary_loss_mlp": 0.01032332,
      "balance_loss_clip": 1.06095994,
      "balance_loss_mlp": 1.02174056,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 2.4894558568024636,
      "language_loss": 0.81065959,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83303374,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.6283438205718994
    },
    {
      "auxiliary_loss_clip": 0.01227706,
      "auxiliary_loss_mlp": 0.01032297,
      "balance_loss_clip": 1.06640911,
      "balance_loss_mlp": 1.02211022,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 3.1316011464498215,
      "language_loss": 0.69322115,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71582115,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.540670871734619
    },
    {
      "auxiliary_loss_clip": 0.01204719,
      "auxiliary_loss_mlp": 0.01040677,
      "balance_loss_clip": 1.06099916,
      "balance_loss_mlp": 1.03044295,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 2.0270127754925573,
      "language_loss": 0.86456645,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88702035,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.7311251163482666
    },
    {
      "auxiliary_loss_clip": 0.01213636,
      "auxiliary_loss_mlp": 0.01043777,
      "balance_loss_clip": 1.06323981,
      "balance_loss_mlp": 1.03329849,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 1.9030972066652696,
      "language_loss": 0.70403779,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.72661191,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.585991144180298
    },
    {
      "auxiliary_loss_clip": 0.01228062,
      "auxiliary_loss_mlp": 0.01036293,
      "balance_loss_clip": 1.06697059,
      "balance_loss_mlp": 1.02590346,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 4.108317712700885,
      "language_loss": 0.94221246,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96485597,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.635230541229248
    },
    {
      "auxiliary_loss_clip": 0.01172021,
      "auxiliary_loss_mlp": 0.01030668,
      "balance_loss_clip": 1.05376148,
      "balance_loss_mlp": 1.02128673,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.069839187003971,
      "language_loss": 0.77936423,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80139112,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.71403169631958
    },
    {
      "auxiliary_loss_clip": 0.01213308,
      "auxiliary_loss_mlp": 0.01036637,
      "balance_loss_clip": 1.06362534,
      "balance_loss_mlp": 1.02617002,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 1.9765153674102878,
      "language_loss": 0.71340859,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73590803,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.614027261734009
    },
    {
      "auxiliary_loss_clip": 0.01205623,
      "auxiliary_loss_mlp": 0.01036356,
      "balance_loss_clip": 1.06295609,
      "balance_loss_mlp": 1.02667594,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 3.948044323467726,
      "language_loss": 0.86276245,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88518226,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.6269893646240234
    },
    {
      "auxiliary_loss_clip": 0.01190559,
      "auxiliary_loss_mlp": 0.01033727,
      "balance_loss_clip": 1.05445933,
      "balance_loss_mlp": 1.02345753,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 2.0439801473469257,
      "language_loss": 0.67842162,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.70066446,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 2.745889663696289
    },
    {
      "auxiliary_loss_clip": 0.01222947,
      "auxiliary_loss_mlp": 0.01040569,
      "balance_loss_clip": 1.06518424,
      "balance_loss_mlp": 1.03136635,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 2.2339707285167942,
      "language_loss": 0.75940311,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78203821,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.549583673477173
    },
    {
      "auxiliary_loss_clip": 0.01182569,
      "auxiliary_loss_mlp": 0.01034524,
      "balance_loss_clip": 1.06068277,
      "balance_loss_mlp": 1.02463603,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 10.916077553267698,
      "language_loss": 0.78837025,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.81054121,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.7571780681610107
    },
    {
      "auxiliary_loss_clip": 0.01211769,
      "auxiliary_loss_mlp": 0.01030143,
      "balance_loss_clip": 1.0611515,
      "balance_loss_mlp": 1.02021849,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.8153986027298323,
      "language_loss": 0.85547435,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87789351,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.658865451812744
    },
    {
      "auxiliary_loss_clip": 0.01172194,
      "auxiliary_loss_mlp": 0.01037645,
      "balance_loss_clip": 1.0501523,
      "balance_loss_mlp": 1.02637982,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.214100064747799,
      "language_loss": 0.81387669,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83597511,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.7442073822021484
    },
    {
      "auxiliary_loss_clip": 0.01212126,
      "auxiliary_loss_mlp": 0.01035106,
      "balance_loss_clip": 1.06163037,
      "balance_loss_mlp": 1.02493739,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 2.4169540230273157,
      "language_loss": 0.85239792,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.87487024,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.6331405639648438
    },
    {
      "auxiliary_loss_clip": 0.01193354,
      "auxiliary_loss_mlp": 0.01033213,
      "balance_loss_clip": 1.05568027,
      "balance_loss_mlp": 1.0224129,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 2.6519269286859495,
      "language_loss": 0.75616848,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77843416,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.623060703277588
    },
    {
      "auxiliary_loss_clip": 0.01208533,
      "auxiliary_loss_mlp": 0.01035299,
      "balance_loss_clip": 1.05839336,
      "balance_loss_mlp": 1.02516055,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 2.3169085719154405,
      "language_loss": 0.73974597,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76218426,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.7174344062805176
    },
    {
      "auxiliary_loss_clip": 0.01190993,
      "auxiliary_loss_mlp": 0.01033369,
      "balance_loss_clip": 1.0563612,
      "balance_loss_mlp": 1.02344513,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 13.397057698445913,
      "language_loss": 0.7983048,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82054842,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.6550791263580322
    },
    {
      "auxiliary_loss_clip": 0.01197091,
      "auxiliary_loss_mlp": 0.00904112,
      "balance_loss_clip": 1.05627608,
      "balance_loss_mlp": 1.00146127,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 2.3351799779387505,
      "language_loss": 0.79366195,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81467396,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.789757251739502
    },
    {
      "auxiliary_loss_clip": 0.01204343,
      "auxiliary_loss_mlp": 0.01035648,
      "balance_loss_clip": 1.0586791,
      "balance_loss_mlp": 1.02600431,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 2.1335630807077304,
      "language_loss": 0.71956897,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.74196887,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 4.554894924163818
    },
    {
      "auxiliary_loss_clip": 0.01211478,
      "auxiliary_loss_mlp": 0.01039699,
      "balance_loss_clip": 1.06137371,
      "balance_loss_mlp": 1.02911365,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.702112189118042,
      "language_loss": 0.7975601,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82007188,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.5903480052948
    },
    {
      "auxiliary_loss_clip": 0.01177988,
      "auxiliary_loss_mlp": 0.01033792,
      "balance_loss_clip": 1.0522095,
      "balance_loss_mlp": 1.02348661,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 3.0452300882204595,
      "language_loss": 0.80195647,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82407427,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.738839864730835
    },
    {
      "auxiliary_loss_clip": 0.01209163,
      "auxiliary_loss_mlp": 0.0103836,
      "balance_loss_clip": 1.06300473,
      "balance_loss_mlp": 1.02764285,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 1.850447161954304,
      "language_loss": 0.65057707,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.67305231,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 4.520962238311768
    },
    {
      "auxiliary_loss_clip": 0.01181219,
      "auxiliary_loss_mlp": 0.01031731,
      "balance_loss_clip": 1.05380595,
      "balance_loss_mlp": 1.02138424,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.292372598427946,
      "language_loss": 0.81742835,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.83955789,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.6820120811462402
    },
    {
      "auxiliary_loss_clip": 0.01194155,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 1.0583241,
      "balance_loss_mlp": 1.02915597,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 2.364252398160206,
      "language_loss": 0.86237025,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88469946,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.7288832664489746
    },
    {
      "auxiliary_loss_clip": 0.01216457,
      "auxiliary_loss_mlp": 0.01035598,
      "balance_loss_clip": 1.06288004,
      "balance_loss_mlp": 1.02511919,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 2.071217865235647,
      "language_loss": 0.9313879,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95390844,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.6461939811706543
    },
    {
      "auxiliary_loss_clip": 0.01215515,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.06256056,
      "balance_loss_mlp": 1.01899767,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.202592781090412,
      "language_loss": 0.79113269,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81358087,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.606358528137207
    },
    {
      "auxiliary_loss_clip": 0.01143389,
      "auxiliary_loss_mlp": 0.01031848,
      "balance_loss_clip": 1.04859328,
      "balance_loss_mlp": 1.02148294,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 2.0559952596168896,
      "language_loss": 0.80866092,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83041334,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.778972625732422
    },
    {
      "auxiliary_loss_clip": 0.01158052,
      "auxiliary_loss_mlp": 0.01034305,
      "balance_loss_clip": 1.05006301,
      "balance_loss_mlp": 1.02436256,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 2.239705100033923,
      "language_loss": 0.73809862,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76002216,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.919172763824463
    },
    {
      "auxiliary_loss_clip": 0.01104636,
      "auxiliary_loss_mlp": 0.01020401,
      "balance_loss_clip": 1.03184795,
      "balance_loss_mlp": 1.01837468,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.039247692126532,
      "language_loss": 0.71145046,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73270082,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.2560341358184814
    },
    {
      "auxiliary_loss_clip": 0.01181553,
      "auxiliary_loss_mlp": 0.01037212,
      "balance_loss_clip": 1.05411935,
      "balance_loss_mlp": 1.02738369,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.7207596666162897,
      "language_loss": 0.77624434,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79843199,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.6603853702545166
    },
    {
      "auxiliary_loss_clip": 0.01175211,
      "auxiliary_loss_mlp": 0.01039609,
      "balance_loss_clip": 1.05758119,
      "balance_loss_mlp": 1.02898192,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.704879800930612,
      "language_loss": 0.79538459,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81753278,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.712066173553467
    },
    {
      "auxiliary_loss_clip": 0.0110664,
      "auxiliary_loss_mlp": 0.01009382,
      "balance_loss_clip": 1.02756023,
      "balance_loss_mlp": 1.00740325,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7795297074276046,
      "language_loss": 0.54945379,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57061398,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.2984249591827393
    },
    {
      "auxiliary_loss_clip": 0.01167431,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.05091107,
      "balance_loss_mlp": 1.01747298,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.888915387126956,
      "language_loss": 0.76611555,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78807116,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.7394042015075684
    },
    {
      "auxiliary_loss_clip": 0.01194369,
      "auxiliary_loss_mlp": 0.01044367,
      "balance_loss_clip": 1.05742478,
      "balance_loss_mlp": 1.03303647,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 4.46992109383392,
      "language_loss": 0.75455177,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77693909,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.728605031967163
    },
    {
      "auxiliary_loss_clip": 0.01214708,
      "auxiliary_loss_mlp": 0.01043208,
      "balance_loss_clip": 1.0615499,
      "balance_loss_mlp": 1.03256822,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 3.6673461504617717,
      "language_loss": 0.76008928,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78266841,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 2.628086566925049
    },
    {
      "auxiliary_loss_clip": 0.01205805,
      "auxiliary_loss_mlp": 0.01040133,
      "balance_loss_clip": 1.06420207,
      "balance_loss_mlp": 1.02981532,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 2.119599428805246,
      "language_loss": 0.74110609,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76356542,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.669982671737671
    },
    {
      "auxiliary_loss_clip": 0.01210364,
      "auxiliary_loss_mlp": 0.01041788,
      "balance_loss_clip": 1.05895805,
      "balance_loss_mlp": 1.03093958,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.5736991081127583,
      "language_loss": 0.76258951,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78511107,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.6366915702819824
    },
    {
      "auxiliary_loss_clip": 0.01218248,
      "auxiliary_loss_mlp": 0.01040325,
      "balance_loss_clip": 1.06142414,
      "balance_loss_mlp": 1.02997756,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.7918819053885433,
      "language_loss": 0.77301157,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79559731,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.5997188091278076
    },
    {
      "auxiliary_loss_clip": 0.01199308,
      "auxiliary_loss_mlp": 0.01041929,
      "balance_loss_clip": 1.06019974,
      "balance_loss_mlp": 1.03098583,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 8.280961249458327,
      "language_loss": 0.79211903,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81453139,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.661755084991455
    },
    {
      "auxiliary_loss_clip": 0.01178378,
      "auxiliary_loss_mlp": 0.01043837,
      "balance_loss_clip": 1.05474257,
      "balance_loss_mlp": 1.0323689,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.140775168001686,
      "language_loss": 0.85860187,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88082403,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.7771708965301514
    },
    {
      "auxiliary_loss_clip": 0.01218333,
      "auxiliary_loss_mlp": 0.00904129,
      "balance_loss_clip": 1.06340504,
      "balance_loss_mlp": 1.001369,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 2.1536542676140034,
      "language_loss": 0.77315056,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79437518,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.616966724395752
    },
    {
      "auxiliary_loss_clip": 0.01213952,
      "auxiliary_loss_mlp": 0.01043528,
      "balance_loss_clip": 1.06532192,
      "balance_loss_mlp": 1.03340077,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 2.141262182286279,
      "language_loss": 0.83217371,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85474855,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.617917537689209
    },
    {
      "auxiliary_loss_clip": 0.01219681,
      "auxiliary_loss_mlp": 0.01038323,
      "balance_loss_clip": 1.06221569,
      "balance_loss_mlp": 1.02801156,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.3465563421247544,
      "language_loss": 0.85022187,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.8728019,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.6091837882995605
    },
    {
      "auxiliary_loss_clip": 0.01199379,
      "auxiliary_loss_mlp": 0.01035857,
      "balance_loss_clip": 1.05365825,
      "balance_loss_mlp": 1.02452636,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.8336074346653175,
      "language_loss": 0.79049975,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81285214,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.6502432823181152
    },
    {
      "auxiliary_loss_clip": 0.01207262,
      "auxiliary_loss_mlp": 0.01038556,
      "balance_loss_clip": 1.05672169,
      "balance_loss_mlp": 1.0283215,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 2.6530124862882607,
      "language_loss": 0.80322933,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82568753,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 4.55270791053772
    },
    {
      "auxiliary_loss_clip": 0.01202384,
      "auxiliary_loss_mlp": 0.01037289,
      "balance_loss_clip": 1.05862594,
      "balance_loss_mlp": 1.02639937,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.8490904279121487,
      "language_loss": 0.75716805,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.77956474,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.7584950923919678
    },
    {
      "auxiliary_loss_clip": 0.01179428,
      "auxiliary_loss_mlp": 0.01040656,
      "balance_loss_clip": 1.05120444,
      "balance_loss_mlp": 1.02980185,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.3973858371427514,
      "language_loss": 0.87505698,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.8972578,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.7265117168426514
    },
    {
      "auxiliary_loss_clip": 0.01196578,
      "auxiliary_loss_mlp": 0.00905304,
      "balance_loss_clip": 1.06261683,
      "balance_loss_mlp": 1.00136948,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.8730320268176108,
      "language_loss": 0.78961772,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.81063652,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 3.5680387020111084
    },
    {
      "auxiliary_loss_clip": 0.01166968,
      "auxiliary_loss_mlp": 0.01036063,
      "balance_loss_clip": 1.05300641,
      "balance_loss_mlp": 1.02607918,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.9192782782915778,
      "language_loss": 0.86318177,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.88521206,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 3.7296152114868164
    },
    {
      "auxiliary_loss_clip": 0.01195828,
      "auxiliary_loss_mlp": 0.0103668,
      "balance_loss_clip": 1.05422997,
      "balance_loss_mlp": 1.02546835,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 2.3071989734448195,
      "language_loss": 0.80390465,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.82622969,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.6890485286712646
    },
    {
      "auxiliary_loss_clip": 0.01204544,
      "auxiliary_loss_mlp": 0.01043255,
      "balance_loss_clip": 1.05966401,
      "balance_loss_mlp": 1.03232908,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 2.6955990932536715,
      "language_loss": 0.74584442,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76832235,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.6748311519622803
    },
    {
      "auxiliary_loss_clip": 0.01217596,
      "auxiliary_loss_mlp": 0.01032381,
      "balance_loss_clip": 1.0645957,
      "balance_loss_mlp": 1.02248669,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 1.9352233518572801,
      "language_loss": 0.86014038,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88264012,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.686683177947998
    },
    {
      "auxiliary_loss_clip": 0.01208417,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.05957913,
      "balance_loss_mlp": 1.02217221,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.4615847144490592,
      "language_loss": 0.83430386,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.8567186,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.614436149597168
    },
    {
      "auxiliary_loss_clip": 0.01219428,
      "auxiliary_loss_mlp": 0.01037084,
      "balance_loss_clip": 1.06122375,
      "balance_loss_mlp": 1.02582479,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 8.13439681704039,
      "language_loss": 0.86411101,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88667613,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.682481527328491
    },
    {
      "auxiliary_loss_clip": 0.01172693,
      "auxiliary_loss_mlp": 0.01041742,
      "balance_loss_clip": 1.05265021,
      "balance_loss_mlp": 1.03150761,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.8011688252701825,
      "language_loss": 0.72847164,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75061607,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.80269455909729
    },
    {
      "auxiliary_loss_clip": 0.01222569,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.06322694,
      "balance_loss_mlp": 1.02471232,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 3.1541068225384215,
      "language_loss": 0.76045322,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78303552,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.671572208404541
    },
    {
      "auxiliary_loss_clip": 0.01210021,
      "auxiliary_loss_mlp": 0.00903923,
      "balance_loss_clip": 1.06199265,
      "balance_loss_mlp": 1.00141573,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 2.865660542099064,
      "language_loss": 0.82902116,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.85016066,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.6232047080993652
    },
    {
      "auxiliary_loss_clip": 0.01213533,
      "auxiliary_loss_mlp": 0.01045287,
      "balance_loss_clip": 1.06121159,
      "balance_loss_mlp": 1.03400373,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.925001529238626,
      "language_loss": 0.73467636,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75726461,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.6806633472442627
    },
    {
      "auxiliary_loss_clip": 0.01179224,
      "auxiliary_loss_mlp": 0.01040084,
      "balance_loss_clip": 1.05765915,
      "balance_loss_mlp": 1.0295161,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.409763401046559,
      "language_loss": 0.88450909,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90670216,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.754889726638794
    },
    {
      "auxiliary_loss_clip": 0.01109747,
      "auxiliary_loss_mlp": 0.01006229,
      "balance_loss_clip": 1.02787757,
      "balance_loss_mlp": 1.00435758,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7661743418915731,
      "language_loss": 0.58196759,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60312736,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.2786247730255127
    },
    {
      "auxiliary_loss_clip": 0.01204383,
      "auxiliary_loss_mlp": 0.01040348,
      "balance_loss_clip": 1.05932403,
      "balance_loss_mlp": 1.02924347,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 3.912034898750941,
      "language_loss": 0.88277555,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90522283,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.6589207649230957
    },
    {
      "auxiliary_loss_clip": 0.0119707,
      "auxiliary_loss_mlp": 0.01035035,
      "balance_loss_clip": 1.06062698,
      "balance_loss_mlp": 1.02428842,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 1.8954430306564929,
      "language_loss": 0.74013436,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.7624554,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.6197197437286377
    },
    {
      "auxiliary_loss_clip": 0.01193363,
      "auxiliary_loss_mlp": 0.01038569,
      "balance_loss_clip": 1.06202078,
      "balance_loss_mlp": 1.02771509,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 2.182834391276324,
      "language_loss": 0.78038394,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80270326,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.9315712451934814
    },
    {
      "auxiliary_loss_clip": 0.01223399,
      "auxiliary_loss_mlp": 0.00904854,
      "balance_loss_clip": 1.06089032,
      "balance_loss_mlp": 1.00147092,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.1291042843767922,
      "language_loss": 0.83963162,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.86091411,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.6423187255859375
    },
    {
      "auxiliary_loss_clip": 0.01211559,
      "auxiliary_loss_mlp": 0.01031574,
      "balance_loss_clip": 1.06168783,
      "balance_loss_mlp": 1.02136946,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.197677581386636,
      "language_loss": 0.84931135,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.87174267,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.613450527191162
    },
    {
      "auxiliary_loss_clip": 0.01219781,
      "auxiliary_loss_mlp": 0.01030836,
      "balance_loss_clip": 1.06201613,
      "balance_loss_mlp": 1.02047729,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 3.3646936480560106,
      "language_loss": 0.78832066,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81082678,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.584468126296997
    },
    {
      "auxiliary_loss_clip": 0.01205849,
      "auxiliary_loss_mlp": 0.00904015,
      "balance_loss_clip": 1.05789626,
      "balance_loss_mlp": 1.00138688,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 2.565126549981725,
      "language_loss": 0.80062509,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.8217237,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.761406660079956
    },
    {
      "auxiliary_loss_clip": 0.01197177,
      "auxiliary_loss_mlp": 0.0103765,
      "balance_loss_clip": 1.06011963,
      "balance_loss_mlp": 1.02691567,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 2.883251755550615,
      "language_loss": 0.82842737,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85077566,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.8054544925689697
    },
    {
      "auxiliary_loss_clip": 0.01190005,
      "auxiliary_loss_mlp": 0.01030956,
      "balance_loss_clip": 1.05614686,
      "balance_loss_mlp": 1.02052569,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 2.0760961400821456,
      "language_loss": 0.87367684,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89588642,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.8079307079315186
    },
    {
      "auxiliary_loss_clip": 0.01199928,
      "auxiliary_loss_mlp": 0.01039573,
      "balance_loss_clip": 1.05871773,
      "balance_loss_mlp": 1.02912462,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 2.6772052443024084,
      "language_loss": 0.71667814,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.73907316,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.699434757232666
    },
    {
      "auxiliary_loss_clip": 0.01197175,
      "auxiliary_loss_mlp": 0.01038172,
      "balance_loss_clip": 1.06187117,
      "balance_loss_mlp": 1.02713943,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 2.1513349128994084,
      "language_loss": 0.74410224,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.76645577,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 4.63434910774231
    },
    {
      "auxiliary_loss_clip": 0.01200656,
      "auxiliary_loss_mlp": 0.01038805,
      "balance_loss_clip": 1.05921996,
      "balance_loss_mlp": 1.02811801,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 3.278622524158848,
      "language_loss": 0.77621943,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.79861403,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.6817026138305664
    },
    {
      "auxiliary_loss_clip": 0.01176867,
      "auxiliary_loss_mlp": 0.01040067,
      "balance_loss_clip": 1.05302608,
      "balance_loss_mlp": 1.02970159,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.9445957915722167,
      "language_loss": 0.85497981,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87714916,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.764394760131836
    },
    {
      "auxiliary_loss_clip": 0.0120843,
      "auxiliary_loss_mlp": 0.01027183,
      "balance_loss_clip": 1.06186092,
      "balance_loss_mlp": 1.01726508,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 2.1398245735079517,
      "language_loss": 0.78873432,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.81109053,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 3.5941998958587646
    },
    {
      "auxiliary_loss_clip": 0.01215413,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.06242239,
      "balance_loss_mlp": 1.01963258,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 3.1907116287054063,
      "language_loss": 0.81820774,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.8406567,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 3.6213111877441406
    },
    {
      "auxiliary_loss_clip": 0.01205526,
      "auxiliary_loss_mlp": 0.01033109,
      "balance_loss_clip": 1.0602051,
      "balance_loss_mlp": 1.02218938,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.512511290294641,
      "language_loss": 0.74136209,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76374853,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.5988292694091797
    },
    {
      "auxiliary_loss_clip": 0.01200877,
      "auxiliary_loss_mlp": 0.00904012,
      "balance_loss_clip": 1.06015921,
      "balance_loss_mlp": 1.00144601,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.8489640110477983,
      "language_loss": 0.78695506,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.80800402,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.703735589981079
    },
    {
      "auxiliary_loss_clip": 0.01088015,
      "auxiliary_loss_mlp": 0.01002537,
      "balance_loss_clip": 1.02985716,
      "balance_loss_mlp": 1.00043929,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.9214306681516148,
      "language_loss": 0.55935812,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58026373,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.274052143096924
    },
    {
      "auxiliary_loss_clip": 0.01192229,
      "auxiliary_loss_mlp": 0.01039039,
      "balance_loss_clip": 1.0571928,
      "balance_loss_mlp": 1.02751732,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 3.4060146309637314,
      "language_loss": 0.79769659,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.82000923,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.845010280609131
    },
    {
      "auxiliary_loss_clip": 0.01201368,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.060004,
      "balance_loss_mlp": 1.02312756,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 3.684957100796641,
      "language_loss": 0.69501042,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.7173574,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.6978204250335693
    },
    {
      "auxiliary_loss_clip": 0.01196129,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.06253791,
      "balance_loss_mlp": 1.02588439,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 1.8999717010723145,
      "language_loss": 0.76940089,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.7917248,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.7449419498443604
    },
    {
      "auxiliary_loss_clip": 0.01117889,
      "auxiliary_loss_mlp": 0.01004574,
      "balance_loss_clip": 1.03360438,
      "balance_loss_mlp": 1.0027504,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.69478197060323,
      "language_loss": 0.58283764,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60406232,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.285055637359619
    },
    {
      "auxiliary_loss_clip": 0.01184134,
      "auxiliary_loss_mlp": 0.01037607,
      "balance_loss_clip": 1.05543423,
      "balance_loss_mlp": 1.02770102,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.8693816747713017,
      "language_loss": 0.76724058,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78945798,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 2.7099568843841553
    },
    {
      "auxiliary_loss_clip": 0.01187517,
      "auxiliary_loss_mlp": 0.01042124,
      "balance_loss_clip": 1.05613637,
      "balance_loss_mlp": 1.03062034,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 10.468956515742832,
      "language_loss": 0.76235533,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78465176,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.6885743141174316
    },
    {
      "auxiliary_loss_clip": 0.01202865,
      "auxiliary_loss_mlp": 0.01037099,
      "balance_loss_clip": 1.05979061,
      "balance_loss_mlp": 1.02678728,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.479772680803515,
      "language_loss": 0.72385097,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74625057,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.715276002883911
    },
    {
      "auxiliary_loss_clip": 0.01225162,
      "auxiliary_loss_mlp": 0.01039393,
      "balance_loss_clip": 1.06346118,
      "balance_loss_mlp": 1.02821088,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 4.120033277965755,
      "language_loss": 0.83727306,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.85991859,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.6011576652526855
    },
    {
      "auxiliary_loss_clip": 0.01199484,
      "auxiliary_loss_mlp": 0.01035478,
      "balance_loss_clip": 1.06063247,
      "balance_loss_mlp": 1.02515399,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 1.6891125993165654,
      "language_loss": 0.79980934,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.82215893,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.6326751708984375
    },
    {
      "auxiliary_loss_clip": 0.01206207,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.0615443,
      "balance_loss_mlp": 1.01882839,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.9695101134476805,
      "language_loss": 0.80771804,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.83006579,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.6297414302825928
    },
    {
      "auxiliary_loss_clip": 0.01225297,
      "auxiliary_loss_mlp": 0.01038275,
      "balance_loss_clip": 1.06391811,
      "balance_loss_mlp": 1.02716446,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 2.382315191956166,
      "language_loss": 0.73298275,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75561845,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.6572916507720947
    },
    {
      "auxiliary_loss_clip": 0.0119988,
      "auxiliary_loss_mlp": 0.0103756,
      "balance_loss_clip": 1.05929804,
      "balance_loss_mlp": 1.02642548,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 1.877501073812605,
      "language_loss": 0.6920138,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71438819,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 2.9374146461486816
    },
    {
      "auxiliary_loss_clip": 0.01220893,
      "auxiliary_loss_mlp": 0.01036957,
      "balance_loss_clip": 1.06269717,
      "balance_loss_mlp": 1.02605486,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 2.6553806788977115,
      "language_loss": 0.86753857,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.89011711,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.6096956729888916
    },
    {
      "auxiliary_loss_clip": 0.01201281,
      "auxiliary_loss_mlp": 0.01042282,
      "balance_loss_clip": 1.05718613,
      "balance_loss_mlp": 1.03136885,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.8225020128739184,
      "language_loss": 0.79448617,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81692177,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.735537052154541
    },
    {
      "auxiliary_loss_clip": 0.01212457,
      "auxiliary_loss_mlp": 0.01041386,
      "balance_loss_clip": 1.06235182,
      "balance_loss_mlp": 1.03027534,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 3.2611470260547133,
      "language_loss": 0.71848512,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74102354,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.6131041049957275
    },
    {
      "auxiliary_loss_clip": 0.01214984,
      "auxiliary_loss_mlp": 0.01032873,
      "balance_loss_clip": 1.06318712,
      "balance_loss_mlp": 1.02256739,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 3.0566871792100847,
      "language_loss": 0.89127564,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91375422,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.6655077934265137
    },
    {
      "auxiliary_loss_clip": 0.01203562,
      "auxiliary_loss_mlp": 0.0103668,
      "balance_loss_clip": 1.0596751,
      "balance_loss_mlp": 1.02685714,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 3.527234688493437,
      "language_loss": 0.68954539,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.7119478,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.6602532863616943
    },
    {
      "auxiliary_loss_clip": 0.01225093,
      "auxiliary_loss_mlp": 0.01038434,
      "balance_loss_clip": 1.06438565,
      "balance_loss_mlp": 1.02713323,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 5.06441602052235,
      "language_loss": 0.84628856,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.8689239,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.575427532196045
    },
    {
      "auxiliary_loss_clip": 0.01224042,
      "auxiliary_loss_mlp": 0.01042019,
      "balance_loss_clip": 1.06552696,
      "balance_loss_mlp": 1.03129661,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 2.311861139323028,
      "language_loss": 0.86594677,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.88860738,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 3.543267011642456
    },
    {
      "auxiliary_loss_clip": 0.01190565,
      "auxiliary_loss_mlp": 0.01032691,
      "balance_loss_clip": 1.05811667,
      "balance_loss_mlp": 1.02210546,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 1.8882293546064315,
      "language_loss": 0.76498973,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.78722227,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 3.6707723140716553
    },
    {
      "auxiliary_loss_clip": 0.01203103,
      "auxiliary_loss_mlp": 0.01035788,
      "balance_loss_clip": 1.06327271,
      "balance_loss_mlp": 1.02525616,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 2.4418022084628777,
      "language_loss": 0.78874213,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81113106,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.6852595806121826
    },
    {
      "auxiliary_loss_clip": 0.01187316,
      "auxiliary_loss_mlp": 0.01049676,
      "balance_loss_clip": 1.05647552,
      "balance_loss_mlp": 1.03704023,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 3.080010288005773,
      "language_loss": 0.82451057,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.8468805,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 3.758239507675171
    },
    {
      "auxiliary_loss_clip": 0.01226833,
      "auxiliary_loss_mlp": 0.01035245,
      "balance_loss_clip": 1.06918585,
      "balance_loss_mlp": 1.02481461,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 2.532706509819414,
      "language_loss": 0.68248332,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70510405,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.612156867980957
    },
    {
      "auxiliary_loss_clip": 0.01203048,
      "auxiliary_loss_mlp": 0.01035335,
      "balance_loss_clip": 1.06004,
      "balance_loss_mlp": 1.02552462,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.896463295879724,
      "language_loss": 0.8865099,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.9088937,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 3.633345603942871
    },
    {
      "auxiliary_loss_clip": 0.0121318,
      "auxiliary_loss_mlp": 0.0104029,
      "balance_loss_clip": 1.06244493,
      "balance_loss_mlp": 1.03018701,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.406152631409585,
      "language_loss": 0.76684773,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.78938246,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.619495391845703
    },
    {
      "auxiliary_loss_clip": 0.01212199,
      "auxiliary_loss_mlp": 0.01038912,
      "balance_loss_clip": 1.06493902,
      "balance_loss_mlp": 1.02910078,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.7795130544971294,
      "language_loss": 0.78917956,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.81169069,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.664430618286133
    },
    {
      "auxiliary_loss_clip": 0.01218984,
      "auxiliary_loss_mlp": 0.01034537,
      "balance_loss_clip": 1.06564736,
      "balance_loss_mlp": 1.02384353,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 3.5458347985166374,
      "language_loss": 0.84953833,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87207347,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.648223638534546
    },
    {
      "auxiliary_loss_clip": 0.01119291,
      "auxiliary_loss_mlp": 0.01009628,
      "balance_loss_clip": 1.03013921,
      "balance_loss_mlp": 1.00756598,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0803118120521824,
      "language_loss": 0.66428506,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68557429,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.210266590118408
    },
    {
      "auxiliary_loss_clip": 0.01188575,
      "auxiliary_loss_mlp": 0.01041092,
      "balance_loss_clip": 1.05909264,
      "balance_loss_mlp": 1.02991629,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.708893996581741,
      "language_loss": 0.69541383,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71771044,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.8499972820281982
    },
    {
      "auxiliary_loss_clip": 0.01109497,
      "auxiliary_loss_mlp": 0.00893252,
      "balance_loss_clip": 1.03428555,
      "balance_loss_mlp": 0.99989855,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 1.0588493680912952,
      "language_loss": 0.61214942,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63217694,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.3729374408721924
    },
    {
      "auxiliary_loss_clip": 0.01226103,
      "auxiliary_loss_mlp": 0.01037735,
      "balance_loss_clip": 1.06620109,
      "balance_loss_mlp": 1.02651739,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 1.9822395313582672,
      "language_loss": 0.77334702,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.7959854,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 2.624572515487671
    },
    {
      "auxiliary_loss_clip": 0.01119551,
      "auxiliary_loss_mlp": 0.010032,
      "balance_loss_clip": 1.02994549,
      "balance_loss_mlp": 1.00135195,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 1.0001332068177287,
      "language_loss": 0.64111531,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66234279,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 2.984890937805176
    },
    {
      "auxiliary_loss_clip": 0.01213016,
      "auxiliary_loss_mlp": 0.01042327,
      "balance_loss_clip": 1.06548882,
      "balance_loss_mlp": 1.03174686,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 3.7742507937302094,
      "language_loss": 0.81000447,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83255786,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.722369432449341
    },
    {
      "auxiliary_loss_clip": 0.01159182,
      "auxiliary_loss_mlp": 0.01031745,
      "balance_loss_clip": 1.05090392,
      "balance_loss_mlp": 1.02180278,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 2.375450696601634,
      "language_loss": 0.82558322,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84749252,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 2.9007418155670166
    },
    {
      "auxiliary_loss_clip": 0.01217363,
      "auxiliary_loss_mlp": 0.01043364,
      "balance_loss_clip": 1.06691742,
      "balance_loss_mlp": 1.03214633,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 2.6739724655774473,
      "language_loss": 0.86268151,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88528883,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.668058395385742
    },
    {
      "auxiliary_loss_clip": 0.01204966,
      "auxiliary_loss_mlp": 0.0090391,
      "balance_loss_clip": 1.06726229,
      "balance_loss_mlp": 1.00133967,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 3.083260798402942,
      "language_loss": 0.80437088,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.8254596,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.699678659439087
    },
    {
      "auxiliary_loss_clip": 0.01215789,
      "auxiliary_loss_mlp": 0.01038363,
      "balance_loss_clip": 1.0649724,
      "balance_loss_mlp": 1.02755117,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.3887988785748697,
      "language_loss": 0.75887859,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78142011,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.652214765548706
    },
    {
      "auxiliary_loss_clip": 0.01196055,
      "auxiliary_loss_mlp": 0.01031823,
      "balance_loss_clip": 1.0566951,
      "balance_loss_mlp": 1.02172577,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.845615358431174,
      "language_loss": 0.86028075,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88255954,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.6867568492889404
    },
    {
      "auxiliary_loss_clip": 0.01193273,
      "auxiliary_loss_mlp": 0.0103072,
      "balance_loss_clip": 1.05835557,
      "balance_loss_mlp": 1.02046204,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 2.2216796625062605,
      "language_loss": 0.75132811,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77356803,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.675808906555176
    },
    {
      "auxiliary_loss_clip": 0.01215819,
      "auxiliary_loss_mlp": 0.01046635,
      "balance_loss_clip": 1.06673622,
      "balance_loss_mlp": 1.03711605,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 2.1905660488931673,
      "language_loss": 0.7552793,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.7779038,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.618321418762207
    },
    {
      "auxiliary_loss_clip": 0.01183308,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.05767417,
      "balance_loss_mlp": 1.02370465,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.8298926529887047,
      "language_loss": 0.74214125,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76431674,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.6807706356048584
    },
    {
      "auxiliary_loss_clip": 0.01225113,
      "auxiliary_loss_mlp": 0.0103587,
      "balance_loss_clip": 1.06636834,
      "balance_loss_mlp": 1.0253855,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.6796716403573755,
      "language_loss": 0.82822788,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85083771,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.6206955909729004
    },
    {
      "auxiliary_loss_clip": 0.01202654,
      "auxiliary_loss_mlp": 0.01036293,
      "balance_loss_clip": 1.06074381,
      "balance_loss_mlp": 1.02511084,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.568332906863545,
      "language_loss": 0.7284646,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.75085413,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.6543567180633545
    },
    {
      "auxiliary_loss_clip": 0.01162653,
      "auxiliary_loss_mlp": 0.01036183,
      "balance_loss_clip": 1.05769205,
      "balance_loss_mlp": 1.02565634,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 1.9294430654991355,
      "language_loss": 0.70961094,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73159933,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.747650146484375
    },
    {
      "auxiliary_loss_clip": 0.01193483,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.06140006,
      "balance_loss_mlp": 1.02577448,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.0429607405705283,
      "language_loss": 0.69099391,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71329254,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 3.6748945713043213
    },
    {
      "auxiliary_loss_clip": 0.01210994,
      "auxiliary_loss_mlp": 0.01032167,
      "balance_loss_clip": 1.06592119,
      "balance_loss_mlp": 1.02285087,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 1.8248167514219147,
      "language_loss": 0.86197817,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88440979,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 3.6208014488220215
    },
    {
      "auxiliary_loss_clip": 0.01099072,
      "auxiliary_loss_mlp": 0.01004311,
      "balance_loss_clip": 1.0236876,
      "balance_loss_mlp": 1.00252271,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.038369473453112,
      "language_loss": 0.62546343,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64649725,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.0813419818878174
    },
    {
      "auxiliary_loss_clip": 0.01230496,
      "auxiliary_loss_mlp": 0.01041603,
      "balance_loss_clip": 1.06740415,
      "balance_loss_mlp": 1.03022456,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 4.056780826057766,
      "language_loss": 0.86553252,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88825351,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.867462158203125
    },
    {
      "auxiliary_loss_clip": 0.01189447,
      "auxiliary_loss_mlp": 0.01039406,
      "balance_loss_clip": 1.06398046,
      "balance_loss_mlp": 1.03016722,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.9946764708396307,
      "language_loss": 0.78272289,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80501139,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 3.7131919860839844
    },
    {
      "auxiliary_loss_clip": 0.01173176,
      "auxiliary_loss_mlp": 0.01043103,
      "balance_loss_clip": 1.05523944,
      "balance_loss_mlp": 1.03214717,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 3.309428894738816,
      "language_loss": 0.77696598,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.79912877,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 3.77413010597229
    },
    {
      "auxiliary_loss_clip": 0.01195578,
      "auxiliary_loss_mlp": 0.00904573,
      "balance_loss_clip": 1.05740345,
      "balance_loss_mlp": 1.0013504,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 2.1545454195205647,
      "language_loss": 0.72580141,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74680293,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 2.7515709400177
    },
    {
      "auxiliary_loss_clip": 0.01194752,
      "auxiliary_loss_mlp": 0.01031557,
      "balance_loss_clip": 1.06147861,
      "balance_loss_mlp": 1.02233016,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.0126978762978944,
      "language_loss": 0.78974253,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.81200564,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.662580728530884
    },
    {
      "auxiliary_loss_clip": 0.01221581,
      "auxiliary_loss_mlp": 0.01038814,
      "balance_loss_clip": 1.06592476,
      "balance_loss_mlp": 1.0289439,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 1.759376904216213,
      "language_loss": 0.85097098,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87357491,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.6448540687561035
    },
    {
      "auxiliary_loss_clip": 0.01189627,
      "auxiliary_loss_mlp": 0.01037081,
      "balance_loss_clip": 1.05483711,
      "balance_loss_mlp": 1.02706707,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 3.0184542707494124,
      "language_loss": 0.78181338,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.80408043,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.6659960746765137
    },
    {
      "auxiliary_loss_clip": 0.01215765,
      "auxiliary_loss_mlp": 0.01030809,
      "balance_loss_clip": 1.06471467,
      "balance_loss_mlp": 1.02178502,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 2.240966786405971,
      "language_loss": 0.71565813,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73812383,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.6863656044006348
    },
    {
      "auxiliary_loss_clip": 0.01209096,
      "auxiliary_loss_mlp": 0.01031676,
      "balance_loss_clip": 1.06415844,
      "balance_loss_mlp": 1.02168679,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.9999158762973874,
      "language_loss": 0.88729435,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90970206,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.6653401851654053
    },
    {
      "auxiliary_loss_clip": 0.01211254,
      "auxiliary_loss_mlp": 0.01040367,
      "balance_loss_clip": 1.06209111,
      "balance_loss_mlp": 1.03019238,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.0225897082154645,
      "language_loss": 0.60212374,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62463999,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 2.7179465293884277
    },
    {
      "auxiliary_loss_clip": 0.01214122,
      "auxiliary_loss_mlp": 0.01041236,
      "balance_loss_clip": 1.06721544,
      "balance_loss_mlp": 1.03135395,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.262911359195931,
      "language_loss": 0.78139496,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.80394852,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.632362127304077
    },
    {
      "auxiliary_loss_clip": 0.01223057,
      "auxiliary_loss_mlp": 0.0103004,
      "balance_loss_clip": 1.06760836,
      "balance_loss_mlp": 1.02007997,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.5764730055929084,
      "language_loss": 0.75511903,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77765,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.611091136932373
    },
    {
      "auxiliary_loss_clip": 0.01201429,
      "auxiliary_loss_mlp": 0.01034392,
      "balance_loss_clip": 1.06126654,
      "balance_loss_mlp": 1.02433062,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.5429175447318733,
      "language_loss": 0.79125112,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81360936,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.7088794708251953
    },
    {
      "auxiliary_loss_clip": 0.01199279,
      "auxiliary_loss_mlp": 0.01037981,
      "balance_loss_clip": 1.0580318,
      "balance_loss_mlp": 1.0288372,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 3.470498557138972,
      "language_loss": 0.8713451,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89371765,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.681407928466797
    },
    {
      "auxiliary_loss_clip": 0.01207491,
      "auxiliary_loss_mlp": 0.01043888,
      "balance_loss_clip": 1.06454206,
      "balance_loss_mlp": 1.03387403,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.0483472458901453,
      "language_loss": 0.76269555,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78520936,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.8018481731414795
    },
    {
      "auxiliary_loss_clip": 0.01203719,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.06112027,
      "balance_loss_mlp": 1.01966023,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.7862350976862524,
      "language_loss": 0.86332536,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88565487,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.7896173000335693
    },
    {
      "auxiliary_loss_clip": 0.01214892,
      "auxiliary_loss_mlp": 0.01033365,
      "balance_loss_clip": 1.06786156,
      "balance_loss_mlp": 1.02394748,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.625164610321209,
      "language_loss": 0.82764351,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.85012603,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.6868205070495605
    },
    {
      "auxiliary_loss_clip": 0.01190029,
      "auxiliary_loss_mlp": 0.01034786,
      "balance_loss_clip": 1.06276774,
      "balance_loss_mlp": 1.02497554,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 2.099552235676399,
      "language_loss": 0.90188915,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92413723,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.6613872051239014
    },
    {
      "auxiliary_loss_clip": 0.01186445,
      "auxiliary_loss_mlp": 0.01038191,
      "balance_loss_clip": 1.0567975,
      "balance_loss_mlp": 1.02795124,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.166144098449718,
      "language_loss": 0.83343351,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85567981,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.6978139877319336
    },
    {
      "auxiliary_loss_clip": 0.01209586,
      "auxiliary_loss_mlp": 0.01037909,
      "balance_loss_clip": 1.06222987,
      "balance_loss_mlp": 1.02790141,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 1.9500940331619274,
      "language_loss": 0.76439226,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78686726,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.6928000450134277
    },
    {
      "auxiliary_loss_clip": 0.01203396,
      "auxiliary_loss_mlp": 0.0104499,
      "balance_loss_clip": 1.06397104,
      "balance_loss_mlp": 1.03469586,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.9131234747850543,
      "language_loss": 0.83522159,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.85770541,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.6590325832366943
    },
    {
      "auxiliary_loss_clip": 0.01164171,
      "auxiliary_loss_mlp": 0.01036294,
      "balance_loss_clip": 1.05163121,
      "balance_loss_mlp": 1.02594638,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 4.343534973566949,
      "language_loss": 0.87170374,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89370847,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.7145235538482666
    },
    {
      "auxiliary_loss_clip": 0.01198111,
      "auxiliary_loss_mlp": 0.01033068,
      "balance_loss_clip": 1.05638945,
      "balance_loss_mlp": 1.02314985,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 2.029185689808196,
      "language_loss": 0.67396539,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69627714,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.739122152328491
    },
    {
      "auxiliary_loss_clip": 0.01192469,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.05880594,
      "balance_loss_mlp": 1.01908958,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.756315649627094,
      "language_loss": 0.74700344,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.7692216,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.6707074642181396
    },
    {
      "auxiliary_loss_clip": 0.01205249,
      "auxiliary_loss_mlp": 0.01034264,
      "balance_loss_clip": 1.06065726,
      "balance_loss_mlp": 1.02438712,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.939139178987454,
      "language_loss": 0.80393481,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82632995,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 4.592056512832642
    },
    {
      "auxiliary_loss_clip": 0.0121288,
      "auxiliary_loss_mlp": 0.01032009,
      "balance_loss_clip": 1.06282544,
      "balance_loss_mlp": 1.02219224,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.9906680583019507,
      "language_loss": 0.82088268,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84333158,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.6559669971466064
    },
    {
      "auxiliary_loss_clip": 0.01198833,
      "auxiliary_loss_mlp": 0.01037745,
      "balance_loss_clip": 1.06355953,
      "balance_loss_mlp": 1.02780914,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 2.289146738005491,
      "language_loss": 0.84453845,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.8669042,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.666280508041382
    },
    {
      "auxiliary_loss_clip": 0.01196653,
      "auxiliary_loss_mlp": 0.01039873,
      "balance_loss_clip": 1.0605408,
      "balance_loss_mlp": 1.02981734,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.857871588636025,
      "language_loss": 0.81917393,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84153914,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.7547659873962402
    },
    {
      "auxiliary_loss_clip": 0.01192454,
      "auxiliary_loss_mlp": 0.01037576,
      "balance_loss_clip": 1.05668306,
      "balance_loss_mlp": 1.02758038,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.3915072740533185,
      "language_loss": 0.78600705,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.80830729,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 3.600992202758789
    },
    {
      "auxiliary_loss_clip": 0.01223968,
      "auxiliary_loss_mlp": 0.01036867,
      "balance_loss_clip": 1.06803465,
      "balance_loss_mlp": 1.02616239,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 2.056312247547931,
      "language_loss": 0.80366409,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82627249,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 3.5583035945892334
    },
    {
      "auxiliary_loss_clip": 0.0117515,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.05479932,
      "balance_loss_mlp": 1.02278495,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 2.181138917443445,
      "language_loss": 0.86026311,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88234663,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.658527374267578
    },
    {
      "auxiliary_loss_clip": 0.0118433,
      "auxiliary_loss_mlp": 0.01041797,
      "balance_loss_clip": 1.05381417,
      "balance_loss_mlp": 1.03112221,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 2.3163381464526194,
      "language_loss": 0.8169778,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.83923906,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.760655164718628
    },
    {
      "auxiliary_loss_clip": 0.01200687,
      "auxiliary_loss_mlp": 0.01032991,
      "balance_loss_clip": 1.06249309,
      "balance_loss_mlp": 1.02267981,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.497436599588653,
      "language_loss": 0.74508047,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.76741725,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.7271742820739746
    },
    {
      "auxiliary_loss_clip": 0.01224287,
      "auxiliary_loss_mlp": 0.01041319,
      "balance_loss_clip": 1.06647706,
      "balance_loss_mlp": 1.03057218,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 1.9203175542520945,
      "language_loss": 0.88034832,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90300441,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.6399738788604736
    },
    {
      "auxiliary_loss_clip": 0.01092631,
      "auxiliary_loss_mlp": 0.01008147,
      "balance_loss_clip": 1.02298617,
      "balance_loss_mlp": 1.00616789,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.875546371856544,
      "language_loss": 0.57923621,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.60024399,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.2672858238220215
    },
    {
      "auxiliary_loss_clip": 0.0122322,
      "auxiliary_loss_mlp": 0.01041065,
      "balance_loss_clip": 1.06581712,
      "balance_loss_mlp": 1.03039014,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 1.9555479288295081,
      "language_loss": 0.75685871,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77950156,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 2.6358330249786377
    },
    {
      "auxiliary_loss_clip": 0.01195558,
      "auxiliary_loss_mlp": 0.01032592,
      "balance_loss_clip": 1.06231952,
      "balance_loss_mlp": 1.02198792,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 14.544428006921573,
      "language_loss": 0.88828111,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91056252,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.6631903648376465
    },
    {
      "auxiliary_loss_clip": 0.01189023,
      "auxiliary_loss_mlp": 0.01037255,
      "balance_loss_clip": 1.05625653,
      "balance_loss_mlp": 1.02679491,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 3.2700876546069493,
      "language_loss": 0.77056408,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79282683,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.664850950241089
    },
    {
      "auxiliary_loss_clip": 0.01195946,
      "auxiliary_loss_mlp": 0.01035619,
      "balance_loss_clip": 1.06183314,
      "balance_loss_mlp": 1.02524149,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.8282402912920448,
      "language_loss": 0.82334912,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84566474,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.695404052734375
    },
    {
      "auxiliary_loss_clip": 0.01108511,
      "auxiliary_loss_mlp": 0.01003252,
      "balance_loss_clip": 1.02591443,
      "balance_loss_mlp": 1.00114238,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7964982310904167,
      "language_loss": 0.59595329,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61707091,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.1184792518615723
    },
    {
      "auxiliary_loss_clip": 0.01179533,
      "auxiliary_loss_mlp": 0.01037471,
      "balance_loss_clip": 1.05684161,
      "balance_loss_mlp": 1.02751708,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 4.47516164577129,
      "language_loss": 0.92977107,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.95194107,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.7957918643951416
    },
    {
      "auxiliary_loss_clip": 0.0121567,
      "auxiliary_loss_mlp": 0.0104143,
      "balance_loss_clip": 1.06655526,
      "balance_loss_mlp": 1.03156543,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 2.3043958923102195,
      "language_loss": 0.88865077,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.9112218,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.6620006561279297
    },
    {
      "auxiliary_loss_clip": 0.01193909,
      "auxiliary_loss_mlp": 0.01044289,
      "balance_loss_clip": 1.06083703,
      "balance_loss_mlp": 1.03290486,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 1.9412085540889696,
      "language_loss": 0.71206981,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73445177,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.7704648971557617
    },
    {
      "auxiliary_loss_clip": 0.01193358,
      "auxiliary_loss_mlp": 0.01044579,
      "balance_loss_clip": 1.0563097,
      "balance_loss_mlp": 1.03276527,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 2.187440398473127,
      "language_loss": 0.78359008,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80596948,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.6925506591796875
    },
    {
      "auxiliary_loss_clip": 0.012126,
      "auxiliary_loss_mlp": 0.0104551,
      "balance_loss_clip": 1.06226563,
      "balance_loss_mlp": 1.0343343,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.56901878675041,
      "language_loss": 0.83758664,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.8601678,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.6548397541046143
    },
    {
      "auxiliary_loss_clip": 0.01199381,
      "auxiliary_loss_mlp": 0.01038997,
      "balance_loss_clip": 1.06116271,
      "balance_loss_mlp": 1.02816129,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 1.813000511644365,
      "language_loss": 0.77915341,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.80153716,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.6552610397338867
    },
    {
      "auxiliary_loss_clip": 0.01216548,
      "auxiliary_loss_mlp": 0.01040303,
      "balance_loss_clip": 1.06619287,
      "balance_loss_mlp": 1.02937722,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.9556734883438307,
      "language_loss": 0.83583498,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85840344,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.6481893062591553
    },
    {
      "auxiliary_loss_clip": 0.01080223,
      "auxiliary_loss_mlp": 0.01003601,
      "balance_loss_clip": 1.02280593,
      "balance_loss_mlp": 1.00151443,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8242105517942172,
      "language_loss": 0.59987688,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62071514,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.1500778198242188
    },
    {
      "auxiliary_loss_clip": 0.01213111,
      "auxiliary_loss_mlp": 0.00904929,
      "balance_loss_clip": 1.06387067,
      "balance_loss_mlp": 1.00148368,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 4.973775884891685,
      "language_loss": 0.84424317,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86542356,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.597102165222168
    },
    {
      "auxiliary_loss_clip": 0.01195841,
      "auxiliary_loss_mlp": 0.00905155,
      "balance_loss_clip": 1.0621202,
      "balance_loss_mlp": 1.00137973,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.7039429547844727,
      "language_loss": 0.86211789,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88312781,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.7831575870513916
    },
    {
      "auxiliary_loss_clip": 0.01218517,
      "auxiliary_loss_mlp": 0.01042271,
      "balance_loss_clip": 1.06623375,
      "balance_loss_mlp": 1.03169131,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.4614270576727595,
      "language_loss": 0.80351377,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82612169,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 3.566861152648926
    },
    {
      "auxiliary_loss_clip": 0.01214613,
      "auxiliary_loss_mlp": 0.01038231,
      "balance_loss_clip": 1.06594503,
      "balance_loss_mlp": 1.02844429,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 3.2612534673558633,
      "language_loss": 0.86920345,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.89173198,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 3.5574543476104736
    },
    {
      "auxiliary_loss_clip": 0.01211204,
      "auxiliary_loss_mlp": 0.01040008,
      "balance_loss_clip": 1.06220508,
      "balance_loss_mlp": 1.02873111,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.0123471289208967,
      "language_loss": 0.8583132,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.88082528,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.6273767948150635
    },
    {
      "auxiliary_loss_clip": 0.01189932,
      "auxiliary_loss_mlp": 0.01035658,
      "balance_loss_clip": 1.05700874,
      "balance_loss_mlp": 1.02392769,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.3467352421288576,
      "language_loss": 0.7612009,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.7834568,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.6852943897247314
    },
    {
      "auxiliary_loss_clip": 0.01109739,
      "auxiliary_loss_mlp": 0.01021142,
      "balance_loss_clip": 1.03326678,
      "balance_loss_mlp": 1.01855516,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8602271464222577,
      "language_loss": 0.61884832,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.64015716,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 4.1587347984313965
    },
    {
      "auxiliary_loss_clip": 0.01220355,
      "auxiliary_loss_mlp": 0.00905289,
      "balance_loss_clip": 1.06363213,
      "balance_loss_mlp": 1.00138199,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 1.7913362283791465,
      "language_loss": 0.83828413,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.85954058,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 3.747387409210205
    },
    {
      "auxiliary_loss_clip": 0.01207744,
      "auxiliary_loss_mlp": 0.01035414,
      "balance_loss_clip": 1.06223738,
      "balance_loss_mlp": 1.0241487,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 2.0749698279971116,
      "language_loss": 0.8617447,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88417625,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.5819172859191895
    },
    {
      "auxiliary_loss_clip": 0.01213509,
      "auxiliary_loss_mlp": 0.01038934,
      "balance_loss_clip": 1.06563497,
      "balance_loss_mlp": 1.02772236,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 2.28984908400371,
      "language_loss": 0.78349769,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80602211,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.7219080924987793
    },
    {
      "auxiliary_loss_clip": 0.0121451,
      "auxiliary_loss_mlp": 0.01035363,
      "balance_loss_clip": 1.06046677,
      "balance_loss_mlp": 1.02502751,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 2.218089903223774,
      "language_loss": 0.84255958,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.8650583,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.5966765880584717
    },
    {
      "auxiliary_loss_clip": 0.0108476,
      "auxiliary_loss_mlp": 0.01004329,
      "balance_loss_clip": 1.02166641,
      "balance_loss_mlp": 1.00232625,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7853040801274082,
      "language_loss": 0.55944729,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58033818,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.131856679916382
    },
    {
      "auxiliary_loss_clip": 0.01183029,
      "auxiliary_loss_mlp": 0.0103209,
      "balance_loss_clip": 1.05471349,
      "balance_loss_mlp": 1.02154624,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 2.0450886440622433,
      "language_loss": 0.8768692,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89902043,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.741389274597168
    },
    {
      "auxiliary_loss_clip": 0.01203349,
      "auxiliary_loss_mlp": 0.01038486,
      "balance_loss_clip": 1.06489003,
      "balance_loss_mlp": 1.02722669,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.3072555667182097,
      "language_loss": 0.7577281,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.78014648,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.635901689529419
    },
    {
      "auxiliary_loss_clip": 0.01188581,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.06278181,
      "balance_loss_mlp": 1.02000928,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 2.5342563704510543,
      "language_loss": 0.82629895,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.84848821,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.7130820751190186
    },
    {
      "auxiliary_loss_clip": 0.01200308,
      "auxiliary_loss_mlp": 0.01039315,
      "balance_loss_clip": 1.06148076,
      "balance_loss_mlp": 1.0287652,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 1.9613602214035237,
      "language_loss": 0.7116574,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.73405367,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.8007848262786865
    },
    {
      "auxiliary_loss_clip": 0.0121527,
      "auxiliary_loss_mlp": 0.01039508,
      "balance_loss_clip": 1.06224525,
      "balance_loss_mlp": 1.02767074,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 2.418991368273413,
      "language_loss": 0.81264901,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83519685,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 2.6514265537261963
    },
    {
      "auxiliary_loss_clip": 0.01209747,
      "auxiliary_loss_mlp": 0.01049417,
      "balance_loss_clip": 1.06454384,
      "balance_loss_mlp": 1.03883147,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.4516974987765814,
      "language_loss": 0.74062639,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76321805,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.724179744720459
    },
    {
      "auxiliary_loss_clip": 0.01202572,
      "auxiliary_loss_mlp": 0.00904726,
      "balance_loss_clip": 1.0614872,
      "balance_loss_mlp": 1.00137711,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 1.9301605822313215,
      "language_loss": 0.79068387,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81175685,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.720674753189087
    },
    {
      "auxiliary_loss_clip": 0.01197125,
      "auxiliary_loss_mlp": 0.01027503,
      "balance_loss_clip": 1.05809796,
      "balance_loss_mlp": 1.0170244,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.372582192778615,
      "language_loss": 0.8118124,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83405876,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.653641700744629
    },
    {
      "auxiliary_loss_clip": 0.01222277,
      "auxiliary_loss_mlp": 0.01034045,
      "balance_loss_clip": 1.06351101,
      "balance_loss_mlp": 1.02279735,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.792744790813113,
      "language_loss": 0.84854817,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.87111139,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.6810288429260254
    },
    {
      "auxiliary_loss_clip": 0.0119097,
      "auxiliary_loss_mlp": 0.01044619,
      "balance_loss_clip": 1.06037045,
      "balance_loss_mlp": 1.03383064,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 2.6873347646446777,
      "language_loss": 0.81735504,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.83971089,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.6802830696105957
    },
    {
      "auxiliary_loss_clip": 0.01183373,
      "auxiliary_loss_mlp": 0.01032731,
      "balance_loss_clip": 1.0554558,
      "balance_loss_mlp": 1.02261651,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 3.52849905818832,
      "language_loss": 0.78666168,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80882275,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.749492883682251
    },
    {
      "auxiliary_loss_clip": 0.0122206,
      "auxiliary_loss_mlp": 0.01039042,
      "balance_loss_clip": 1.06681299,
      "balance_loss_mlp": 1.02884948,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 2.035741913365545,
      "language_loss": 0.80925453,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.83186555,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.559462070465088
    },
    {
      "auxiliary_loss_clip": 0.01203561,
      "auxiliary_loss_mlp": 0.01035769,
      "balance_loss_clip": 1.06004524,
      "balance_loss_mlp": 1.02545762,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.6935533770063693,
      "language_loss": 0.82499766,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.84739095,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.6957736015319824
    },
    {
      "auxiliary_loss_clip": 0.01199408,
      "auxiliary_loss_mlp": 0.01034264,
      "balance_loss_clip": 1.06225157,
      "balance_loss_mlp": 1.02386332,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 5.009940134780209,
      "language_loss": 0.83473045,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85706711,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.6223175525665283
    },
    {
      "auxiliary_loss_clip": 0.01190596,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.05515695,
      "balance_loss_mlp": 1.02193224,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.7942948940674217,
      "language_loss": 0.78479058,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80702364,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.7393314838409424
    },
    {
      "auxiliary_loss_clip": 0.01211535,
      "auxiliary_loss_mlp": 0.01041092,
      "balance_loss_clip": 1.06464696,
      "balance_loss_mlp": 1.03141809,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.4561119946614722,
      "language_loss": 0.77963179,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.802158,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.6291046142578125
    },
    {
      "auxiliary_loss_clip": 0.012023,
      "auxiliary_loss_mlp": 0.01041052,
      "balance_loss_clip": 1.06147718,
      "balance_loss_mlp": 1.02907741,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 2.6236120371506293,
      "language_loss": 0.74940968,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77184325,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.6672184467315674
    },
    {
      "auxiliary_loss_clip": 0.01203451,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.06573975,
      "balance_loss_mlp": 1.02883291,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 2.112531444185446,
      "language_loss": 0.81900561,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.84142864,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 3.609874725341797
    },
    {
      "auxiliary_loss_clip": 0.01185665,
      "auxiliary_loss_mlp": 0.01038722,
      "balance_loss_clip": 1.05743945,
      "balance_loss_mlp": 1.0285778,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 1.7538334193467733,
      "language_loss": 0.74448705,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.7667309,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.649369716644287
    },
    {
      "auxiliary_loss_clip": 0.01202754,
      "auxiliary_loss_mlp": 0.01045539,
      "balance_loss_clip": 1.05782366,
      "balance_loss_mlp": 1.0351615,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 2.971417119192675,
      "language_loss": 0.72664773,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.74913061,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.7212483882904053
    },
    {
      "auxiliary_loss_clip": 0.01226253,
      "auxiliary_loss_mlp": 0.01037478,
      "balance_loss_clip": 1.06473303,
      "balance_loss_mlp": 1.02655256,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 5.824769521461303,
      "language_loss": 0.77556443,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.79820174,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.589317798614502
    },
    {
      "auxiliary_loss_clip": 0.01206675,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.06115103,
      "balance_loss_mlp": 1.02240682,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.509804295271481,
      "language_loss": 0.80238354,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82478821,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 3.552469491958618
    },
    {
      "auxiliary_loss_clip": 0.01201497,
      "auxiliary_loss_mlp": 0.01034377,
      "balance_loss_clip": 1.0655272,
      "balance_loss_mlp": 1.0254004,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 7.6663141772841685,
      "language_loss": 0.80369353,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82605231,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 3.5630953311920166
    },
    {
      "auxiliary_loss_clip": 0.01083792,
      "auxiliary_loss_mlp": 0.01008926,
      "balance_loss_clip": 1.02398252,
      "balance_loss_mlp": 1.00681579,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9391162390403031,
      "language_loss": 0.6320681,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65299523,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.1312596797943115
    },
    {
      "auxiliary_loss_clip": 0.01189974,
      "auxiliary_loss_mlp": 0.01034319,
      "balance_loss_clip": 1.06019521,
      "balance_loss_mlp": 1.02327394,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 2.0366858097743346,
      "language_loss": 0.72079396,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74303699,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 2.772099494934082
    },
    {
      "auxiliary_loss_clip": 0.01204275,
      "auxiliary_loss_mlp": 0.01038339,
      "balance_loss_clip": 1.06148958,
      "balance_loss_mlp": 1.02793229,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 3.4021139729640377,
      "language_loss": 0.82006347,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84248972,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.6571881771087646
    },
    {
      "auxiliary_loss_clip": 0.01222812,
      "auxiliary_loss_mlp": 0.01036892,
      "balance_loss_clip": 1.06474018,
      "balance_loss_mlp": 1.02580523,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.4861133743926387,
      "language_loss": 0.87596452,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.89856154,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.5710232257843018
    },
    {
      "auxiliary_loss_clip": 0.01212233,
      "auxiliary_loss_mlp": 0.01038633,
      "balance_loss_clip": 1.06080568,
      "balance_loss_mlp": 1.02830958,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 3.101299980204936,
      "language_loss": 0.85111672,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.8736254,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.653933048248291
    },
    {
      "auxiliary_loss_clip": 0.01180515,
      "auxiliary_loss_mlp": 0.01039868,
      "balance_loss_clip": 1.05977964,
      "balance_loss_mlp": 1.02813768,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.8279024104472628,
      "language_loss": 0.8000375,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.82224131,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.7734484672546387
    },
    {
      "auxiliary_loss_clip": 0.01213052,
      "auxiliary_loss_mlp": 0.01035178,
      "balance_loss_clip": 1.06676483,
      "balance_loss_mlp": 1.0250814,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 2.0293112996423055,
      "language_loss": 0.8131299,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83561218,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.7013163566589355
    },
    {
      "auxiliary_loss_clip": 0.01177993,
      "auxiliary_loss_mlp": 0.01042602,
      "balance_loss_clip": 1.06311393,
      "balance_loss_mlp": 1.03179598,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 2.4232167716592943,
      "language_loss": 0.80605483,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.82826078,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 2.6898510456085205
    },
    {
      "auxiliary_loss_clip": 0.01206847,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.0628953,
      "balance_loss_mlp": 1.01857114,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.4772606784785283,
      "language_loss": 0.95820296,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.98055828,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.7303237915039062
    },
    {
      "auxiliary_loss_clip": 0.01165668,
      "auxiliary_loss_mlp": 0.01035925,
      "balance_loss_clip": 1.04892612,
      "balance_loss_mlp": 1.02496958,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 2.11444713887325,
      "language_loss": 0.7884084,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.81042433,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.8442511558532715
    },
    {
      "auxiliary_loss_clip": 0.01205328,
      "auxiliary_loss_mlp": 0.01035305,
      "balance_loss_clip": 1.06315541,
      "balance_loss_mlp": 1.02536333,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 4.113629052017617,
      "language_loss": 0.78733671,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80974299,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.7001161575317383
    },
    {
      "auxiliary_loss_clip": 0.01200014,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.06344867,
      "balance_loss_mlp": 1.02219963,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 1.9171185039354774,
      "language_loss": 0.78206354,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80438632,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.677567958831787
    },
    {
      "auxiliary_loss_clip": 0.01212697,
      "auxiliary_loss_mlp": 0.01029989,
      "balance_loss_clip": 1.06214309,
      "balance_loss_mlp": 1.02004147,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.6821189842114888,
      "language_loss": 0.73554182,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.7579686,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.736022710800171
    },
    {
      "auxiliary_loss_clip": 0.01192144,
      "auxiliary_loss_mlp": 0.00905683,
      "balance_loss_clip": 1.0592165,
      "balance_loss_mlp": 1.00140858,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.084399655813564,
      "language_loss": 0.7889275,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.80990577,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.71532940864563
    },
    {
      "auxiliary_loss_clip": 0.01195298,
      "auxiliary_loss_mlp": 0.00904481,
      "balance_loss_clip": 1.06255341,
      "balance_loss_mlp": 1.0015527,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.256266674122855,
      "language_loss": 0.78127873,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80227649,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.704556465148926
    },
    {
      "auxiliary_loss_clip": 0.01094063,
      "auxiliary_loss_mlp": 0.0100375,
      "balance_loss_clip": 1.02481842,
      "balance_loss_mlp": 1.00149679,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7576125935741332,
      "language_loss": 0.59406257,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61504066,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.1698808670043945
    },
    {
      "auxiliary_loss_clip": 0.01222126,
      "auxiliary_loss_mlp": 0.01036904,
      "balance_loss_clip": 1.06634557,
      "balance_loss_mlp": 1.02622306,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 3.175269818923136,
      "language_loss": 0.82830238,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.85089272,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.6201789379119873
    },
    {
      "auxiliary_loss_clip": 0.01173407,
      "auxiliary_loss_mlp": 0.01035258,
      "balance_loss_clip": 1.05945325,
      "balance_loss_mlp": 1.02500618,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.9660566241316884,
      "language_loss": 0.83161998,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85370666,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.7975757122039795
    },
    {
      "auxiliary_loss_clip": 0.01218707,
      "auxiliary_loss_mlp": 0.01035042,
      "balance_loss_clip": 1.06487083,
      "balance_loss_mlp": 1.02477217,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.165719628443929,
      "language_loss": 0.89906734,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92160487,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.673616886138916
    },
    {
      "auxiliary_loss_clip": 0.01210925,
      "auxiliary_loss_mlp": 0.01038173,
      "balance_loss_clip": 1.06142294,
      "balance_loss_mlp": 1.02798629,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 1.9661824368802923,
      "language_loss": 0.80832791,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.83081889,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.7595033645629883
    },
    {
      "auxiliary_loss_clip": 0.01206768,
      "auxiliary_loss_mlp": 0.01035696,
      "balance_loss_clip": 1.06291175,
      "balance_loss_mlp": 1.02565908,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.6644561032345087,
      "language_loss": 0.72763819,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75006282,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 3.5939443111419678
    },
    {
      "auxiliary_loss_clip": 0.01204327,
      "auxiliary_loss_mlp": 0.01033966,
      "balance_loss_clip": 1.06047058,
      "balance_loss_mlp": 1.02348137,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.2370023894702014,
      "language_loss": 0.75533545,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77771842,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 3.4897286891937256
    },
    {
      "auxiliary_loss_clip": 0.01219579,
      "auxiliary_loss_mlp": 0.010435,
      "balance_loss_clip": 1.06376123,
      "balance_loss_mlp": 1.03368878,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 2.419068199666279,
      "language_loss": 0.82275987,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84539068,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.68575119972229
    },
    {
      "auxiliary_loss_clip": 0.0121395,
      "auxiliary_loss_mlp": 0.01041455,
      "balance_loss_clip": 1.06562161,
      "balance_loss_mlp": 1.02993953,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 4.015532245669689,
      "language_loss": 0.80166364,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82421768,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.550060272216797
    },
    {
      "auxiliary_loss_clip": 0.01207385,
      "auxiliary_loss_mlp": 0.01033413,
      "balance_loss_clip": 1.06385338,
      "balance_loss_mlp": 1.02277946,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 6.9770766573105645,
      "language_loss": 0.84362668,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86603463,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 3.5556111335754395
    },
    {
      "auxiliary_loss_clip": 0.01162737,
      "auxiliary_loss_mlp": 0.01033077,
      "balance_loss_clip": 1.05460918,
      "balance_loss_mlp": 1.02276576,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 1.9164568155647628,
      "language_loss": 0.8246944,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.84665251,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.786604404449463
    },
    {
      "auxiliary_loss_clip": 0.01193458,
      "auxiliary_loss_mlp": 0.01041868,
      "balance_loss_clip": 1.06466889,
      "balance_loss_mlp": 1.0306983,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.7091989630105835,
      "language_loss": 0.75267136,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77502459,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 3.675720453262329
    },
    {
      "auxiliary_loss_clip": 0.01218267,
      "auxiliary_loss_mlp": 0.01034999,
      "balance_loss_clip": 1.06615031,
      "balance_loss_mlp": 1.02518773,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.668322425094742,
      "language_loss": 0.78563595,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80816853,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 2.7735915184020996
    },
    {
      "auxiliary_loss_clip": 0.01190666,
      "auxiliary_loss_mlp": 0.01034759,
      "balance_loss_clip": 1.0607264,
      "balance_loss_mlp": 1.02426815,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.8476830278738638,
      "language_loss": 0.8075245,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.82977867,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.8379435539245605
    },
    {
      "auxiliary_loss_clip": 0.01207059,
      "auxiliary_loss_mlp": 0.01042113,
      "balance_loss_clip": 1.06210327,
      "balance_loss_mlp": 1.03220117,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.050874982708828,
      "language_loss": 0.76926064,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79175234,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.676844835281372
    },
    {
      "auxiliary_loss_clip": 0.01191702,
      "auxiliary_loss_mlp": 0.01035553,
      "balance_loss_clip": 1.05650461,
      "balance_loss_mlp": 1.02511644,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.8921686283133938,
      "language_loss": 0.81382895,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83610153,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.718764543533325
    },
    {
      "auxiliary_loss_clip": 0.01222111,
      "auxiliary_loss_mlp": 0.01038756,
      "balance_loss_clip": 1.0679493,
      "balance_loss_mlp": 1.02828348,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 3.6420371884341036,
      "language_loss": 0.76370835,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78631699,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.6306777000427246
    },
    {
      "auxiliary_loss_clip": 0.01201375,
      "auxiliary_loss_mlp": 0.01042512,
      "balance_loss_clip": 1.06084418,
      "balance_loss_mlp": 1.03193784,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.535674707200055,
      "language_loss": 0.84061003,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.86304885,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 2.7886476516723633
    },
    {
      "auxiliary_loss_clip": 0.01147303,
      "auxiliary_loss_mlp": 0.01029644,
      "balance_loss_clip": 1.05001295,
      "balance_loss_mlp": 1.01951146,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 2.013008428399102,
      "language_loss": 0.80928528,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83105475,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 2.8298180103302
    },
    {
      "auxiliary_loss_clip": 0.01220533,
      "auxiliary_loss_mlp": 0.0103484,
      "balance_loss_clip": 1.06643629,
      "balance_loss_mlp": 1.02432537,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.234117672999409,
      "language_loss": 0.85020489,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87275863,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.653451442718506
    },
    {
      "auxiliary_loss_clip": 0.01204334,
      "auxiliary_loss_mlp": 0.01037209,
      "balance_loss_clip": 1.06101954,
      "balance_loss_mlp": 1.02634871,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.6970332300624984,
      "language_loss": 0.80364507,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82606047,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.6590969562530518
    },
    {
      "auxiliary_loss_clip": 0.01204638,
      "auxiliary_loss_mlp": 0.01033041,
      "balance_loss_clip": 1.06446838,
      "balance_loss_mlp": 1.02265215,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 1.9916832655652792,
      "language_loss": 0.7695089,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79188573,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.7569258213043213
    },
    {
      "auxiliary_loss_clip": 0.01171758,
      "auxiliary_loss_mlp": 0.01034145,
      "balance_loss_clip": 1.05696797,
      "balance_loss_mlp": 1.02283251,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 2.111640287067388,
      "language_loss": 0.76459837,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78665745,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.731837034225464
    },
    {
      "auxiliary_loss_clip": 0.01222014,
      "auxiliary_loss_mlp": 0.00904762,
      "balance_loss_clip": 1.06720018,
      "balance_loss_mlp": 1.00178719,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.1869859838009935,
      "language_loss": 0.81896442,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.84023219,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.634861707687378
    },
    {
      "auxiliary_loss_clip": 0.01182593,
      "auxiliary_loss_mlp": 0.01038959,
      "balance_loss_clip": 1.05580378,
      "balance_loss_mlp": 1.02957129,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 2.225889942636526,
      "language_loss": 0.78095859,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80317414,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.8029842376708984
    },
    {
      "auxiliary_loss_clip": 0.01213051,
      "auxiliary_loss_mlp": 0.01037893,
      "balance_loss_clip": 1.06251884,
      "balance_loss_mlp": 1.0279274,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 5.086901842296429,
      "language_loss": 0.85128272,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87379217,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.6409049034118652
    },
    {
      "auxiliary_loss_clip": 0.01092616,
      "auxiliary_loss_mlp": 0.01003852,
      "balance_loss_clip": 1.02762425,
      "balance_loss_mlp": 1.00190914,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.7169574403327709,
      "language_loss": 0.59124458,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.6122092,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.204557418823242
    },
    {
      "auxiliary_loss_clip": 0.01205536,
      "auxiliary_loss_mlp": 0.01031728,
      "balance_loss_clip": 1.06470537,
      "balance_loss_mlp": 1.02173793,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 5.065638054582952,
      "language_loss": 0.65012407,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67249674,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.709247589111328
    },
    {
      "auxiliary_loss_clip": 0.01187974,
      "auxiliary_loss_mlp": 0.0103861,
      "balance_loss_clip": 1.06350553,
      "balance_loss_mlp": 1.02740455,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 4.352757173576933,
      "language_loss": 0.64925247,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67151833,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.723341703414917
    },
    {
      "auxiliary_loss_clip": 0.01192623,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 1.06222236,
      "balance_loss_mlp": 1.02202034,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 2.3312000060689715,
      "language_loss": 0.89097404,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91321784,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.659020185470581
    },
    {
      "auxiliary_loss_clip": 0.01175873,
      "auxiliary_loss_mlp": 0.01045031,
      "balance_loss_clip": 1.05589771,
      "balance_loss_mlp": 1.03542876,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.3794438663009556,
      "language_loss": 0.90552104,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92773008,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.759244918823242
    },
    {
      "auxiliary_loss_clip": 0.01212078,
      "auxiliary_loss_mlp": 0.01044611,
      "balance_loss_clip": 1.06137335,
      "balance_loss_mlp": 1.03406096,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 3.2759539206609984,
      "language_loss": 0.70144576,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72401261,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.655921459197998
    },
    {
      "auxiliary_loss_clip": 0.01204731,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.05998921,
      "balance_loss_mlp": 1.01887834,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 3.383528926104867,
      "language_loss": 0.81577438,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83810747,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 4.570174932479858
    },
    {
      "auxiliary_loss_clip": 0.0120462,
      "auxiliary_loss_mlp": 0.01040344,
      "balance_loss_clip": 1.06940663,
      "balance_loss_mlp": 1.03002024,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 2.370928152071625,
      "language_loss": 0.77710581,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79955548,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.6793763637542725
    },
    {
      "auxiliary_loss_clip": 0.01189003,
      "auxiliary_loss_mlp": 0.01038416,
      "balance_loss_clip": 1.0547781,
      "balance_loss_mlp": 1.02890944,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.5296346170925177,
      "language_loss": 0.80599332,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.82826746,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.722747564315796
    },
    {
      "auxiliary_loss_clip": 0.01208809,
      "auxiliary_loss_mlp": 0.01035048,
      "balance_loss_clip": 1.06441736,
      "balance_loss_mlp": 1.02507663,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 2.2575999483468268,
      "language_loss": 0.81300098,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83543956,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 3.6015641689300537
    },
    {
      "auxiliary_loss_clip": 0.01112048,
      "auxiliary_loss_mlp": 0.01001293,
      "balance_loss_clip": 1.0304997,
      "balance_loss_mlp": 0.999255,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8323943858067271,
      "language_loss": 0.61055756,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63169098,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.1938893795013428
    },
    {
      "auxiliary_loss_clip": 0.01189241,
      "auxiliary_loss_mlp": 0.01035979,
      "balance_loss_clip": 1.06186652,
      "balance_loss_mlp": 1.02630472,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.5877102566752932,
      "language_loss": 0.75695354,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77920568,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.747939109802246
    },
    {
      "auxiliary_loss_clip": 0.01191896,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 1.05909228,
      "balance_loss_mlp": 1.02244842,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 2.145711612985308,
      "language_loss": 0.77782863,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.80007243,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 3.637277364730835
    },
    {
      "auxiliary_loss_clip": 0.01187536,
      "auxiliary_loss_mlp": 0.01032475,
      "balance_loss_clip": 1.05565226,
      "balance_loss_mlp": 1.02185988,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 3.4929597507330894,
      "language_loss": 0.77025902,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.79245913,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 2.6670427322387695
    },
    {
      "auxiliary_loss_clip": 0.0119605,
      "auxiliary_loss_mlp": 0.01035928,
      "balance_loss_clip": 1.05894208,
      "balance_loss_mlp": 1.02654016,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 2.079963116063347,
      "language_loss": 0.82627535,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.84859514,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.652418375015259
    },
    {
      "auxiliary_loss_clip": 0.01191721,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.05955648,
      "balance_loss_mlp": 1.02400661,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 1.8118427098284116,
      "language_loss": 0.87300295,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89525771,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.7313802242279053
    },
    {
      "auxiliary_loss_clip": 0.01217934,
      "auxiliary_loss_mlp": 0.01038551,
      "balance_loss_clip": 1.06338894,
      "balance_loss_mlp": 1.02858543,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 2.874653651961056,
      "language_loss": 0.75687802,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.77944279,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.665278196334839
    },
    {
      "auxiliary_loss_clip": 0.01177805,
      "auxiliary_loss_mlp": 0.00904359,
      "balance_loss_clip": 1.05811799,
      "balance_loss_mlp": 1.00207889,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.1563746862369184,
      "language_loss": 0.8640185,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88484013,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 2.8276610374450684
    },
    {
      "auxiliary_loss_clip": 0.01199353,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.05850995,
      "balance_loss_mlp": 1.02151799,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 1.9797118599188541,
      "language_loss": 0.83232492,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.8546344,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.5880885124206543
    },
    {
      "auxiliary_loss_clip": 0.0117859,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 1.05419469,
      "balance_loss_mlp": 1.0248847,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 3.1262787047434855,
      "language_loss": 0.75829595,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.78043073,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.7754476070404053
    },
    {
      "auxiliary_loss_clip": 0.01194213,
      "auxiliary_loss_mlp": 0.01034979,
      "balance_loss_clip": 1.05657101,
      "balance_loss_mlp": 1.02503133,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.7229559360330213,
      "language_loss": 0.81611645,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83840847,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.7965316772460938
    },
    {
      "auxiliary_loss_clip": 0.01184444,
      "auxiliary_loss_mlp": 0.01034187,
      "balance_loss_clip": 1.06042337,
      "balance_loss_mlp": 1.02408361,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.481937309242556,
      "language_loss": 0.77445459,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79664087,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.7507126331329346
    },
    {
      "auxiliary_loss_clip": 0.01194309,
      "auxiliary_loss_mlp": 0.01036569,
      "balance_loss_clip": 1.05701113,
      "balance_loss_mlp": 1.02715743,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 3.3460844221544983,
      "language_loss": 0.85970217,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88201094,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.756164312362671
    },
    {
      "auxiliary_loss_clip": 0.01186653,
      "auxiliary_loss_mlp": 0.01032294,
      "balance_loss_clip": 1.05517578,
      "balance_loss_mlp": 1.02269721,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.4047575773091276,
      "language_loss": 0.82652348,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.84871292,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.7128260135650635
    },
    {
      "auxiliary_loss_clip": 0.01196826,
      "auxiliary_loss_mlp": 0.01036617,
      "balance_loss_clip": 1.05832982,
      "balance_loss_mlp": 1.02736664,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.1594219414730262,
      "language_loss": 0.78003502,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80236948,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.650242567062378
    },
    {
      "auxiliary_loss_clip": 0.01212684,
      "auxiliary_loss_mlp": 0.01036218,
      "balance_loss_clip": 1.06738806,
      "balance_loss_mlp": 1.02713418,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.8800525564406096,
      "language_loss": 0.86898792,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.89147693,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.632650136947632
    },
    {
      "auxiliary_loss_clip": 0.01169907,
      "auxiliary_loss_mlp": 0.01037104,
      "balance_loss_clip": 1.05022788,
      "balance_loss_mlp": 1.02794862,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.254807983770694,
      "language_loss": 0.71191311,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73398328,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.7004053592681885
    },
    {
      "auxiliary_loss_clip": 0.01206175,
      "auxiliary_loss_mlp": 0.01037368,
      "balance_loss_clip": 1.06498408,
      "balance_loss_mlp": 1.02746749,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 5.292027467678455,
      "language_loss": 0.80826724,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83070266,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.757413148880005
    },
    {
      "auxiliary_loss_clip": 0.01221657,
      "auxiliary_loss_mlp": 0.01042011,
      "balance_loss_clip": 1.0657351,
      "balance_loss_mlp": 1.03152633,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 2.100012579462213,
      "language_loss": 0.68856215,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71119887,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.6458089351654053
    },
    {
      "auxiliary_loss_clip": 0.01212872,
      "auxiliary_loss_mlp": 0.01033092,
      "balance_loss_clip": 1.06087971,
      "balance_loss_mlp": 1.02302516,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 3.6168756355256386,
      "language_loss": 0.65900803,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68146771,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.7435176372528076
    },
    {
      "auxiliary_loss_clip": 0.01213922,
      "auxiliary_loss_mlp": 0.01039412,
      "balance_loss_clip": 1.06281173,
      "balance_loss_mlp": 1.02907658,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.9483056374313037,
      "language_loss": 0.86117673,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88371003,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.57769513130188
    },
    {
      "auxiliary_loss_clip": 0.01201153,
      "auxiliary_loss_mlp": 0.01037237,
      "balance_loss_clip": 1.0609349,
      "balance_loss_mlp": 1.02823102,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.1866242795646467,
      "language_loss": 0.78411847,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.8065024,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.617816925048828
    },
    {
      "auxiliary_loss_clip": 0.01180283,
      "auxiliary_loss_mlp": 0.01039886,
      "balance_loss_clip": 1.05299044,
      "balance_loss_mlp": 1.02855551,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.3139556454843695,
      "language_loss": 0.75134766,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77354938,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 3.590226650238037
    },
    {
      "auxiliary_loss_clip": 0.01197798,
      "auxiliary_loss_mlp": 0.01033296,
      "balance_loss_clip": 1.05779886,
      "balance_loss_mlp": 1.02389657,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 2.0953637045786424,
      "language_loss": 0.87595189,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89826286,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 3.6096291542053223
    },
    {
      "auxiliary_loss_clip": 0.01170647,
      "auxiliary_loss_mlp": 0.01031569,
      "balance_loss_clip": 1.05336416,
      "balance_loss_mlp": 1.02210379,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 2.47099156008956,
      "language_loss": 0.70427084,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.72629297,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.694077491760254
    },
    {
      "auxiliary_loss_clip": 0.01188531,
      "auxiliary_loss_mlp": 0.01035683,
      "balance_loss_clip": 1.05566335,
      "balance_loss_mlp": 1.02602136,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 1.9940793946480055,
      "language_loss": 0.91544843,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.93769062,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 3.6227867603302
    },
    {
      "auxiliary_loss_clip": 0.01208848,
      "auxiliary_loss_mlp": 0.00904421,
      "balance_loss_clip": 1.06020117,
      "balance_loss_mlp": 1.00217295,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 2.0181057520268935,
      "language_loss": 0.79721582,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.81834859,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.6136226654052734
    },
    {
      "auxiliary_loss_clip": 0.01201865,
      "auxiliary_loss_mlp": 0.01031268,
      "balance_loss_clip": 1.0581708,
      "balance_loss_mlp": 1.02171302,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.3456161925880252,
      "language_loss": 0.84349722,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.86582851,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.6442711353302
    },
    {
      "auxiliary_loss_clip": 0.01173709,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.05579865,
      "balance_loss_mlp": 1.02488363,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.515654084854177,
      "language_loss": 0.77438784,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79647207,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.7241690158843994
    },
    {
      "auxiliary_loss_clip": 0.01185333,
      "auxiliary_loss_mlp": 0.01037577,
      "balance_loss_clip": 1.05708122,
      "balance_loss_mlp": 1.02650237,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 3.9156440015313505,
      "language_loss": 0.80908358,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83131266,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 3.5917396545410156
    },
    {
      "auxiliary_loss_clip": 0.01192287,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 1.05642962,
      "balance_loss_mlp": 1.02361941,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.7495386708402083,
      "language_loss": 0.74774492,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.7699948,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.7220892906188965
    },
    {
      "auxiliary_loss_clip": 0.01213187,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.0611136,
      "balance_loss_mlp": 1.02380764,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 3.1616780742824404,
      "language_loss": 0.81345081,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83591866,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.5548818111419678
    },
    {
      "auxiliary_loss_clip": 0.01197629,
      "auxiliary_loss_mlp": 0.00904327,
      "balance_loss_clip": 1.06033123,
      "balance_loss_mlp": 1.00209856,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 2.133947666074446,
      "language_loss": 0.81323212,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83425164,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.7914299964904785
    },
    {
      "auxiliary_loss_clip": 0.01202612,
      "auxiliary_loss_mlp": 0.01037175,
      "balance_loss_clip": 1.05693042,
      "balance_loss_mlp": 1.02746582,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 2.0419563675630528,
      "language_loss": 0.83565158,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85804945,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.6195032596588135
    },
    {
      "auxiliary_loss_clip": 0.01212063,
      "auxiliary_loss_mlp": 0.01036307,
      "balance_loss_clip": 1.059389,
      "balance_loss_mlp": 1.0265913,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 7.098668453793624,
      "language_loss": 0.85108078,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87356442,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.6485447883605957
    },
    {
      "auxiliary_loss_clip": 0.01212269,
      "auxiliary_loss_mlp": 0.01034903,
      "balance_loss_clip": 1.06382477,
      "balance_loss_mlp": 1.02572954,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.9186170754716194,
      "language_loss": 0.79763663,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.82010841,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 2.61511492729187
    },
    {
      "auxiliary_loss_clip": 0.01197712,
      "auxiliary_loss_mlp": 0.01033532,
      "balance_loss_clip": 1.05753827,
      "balance_loss_mlp": 1.02485371,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.8863264911106432,
      "language_loss": 0.87546921,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.89778161,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.6616528034210205
    },
    {
      "auxiliary_loss_clip": 0.0120092,
      "auxiliary_loss_mlp": 0.01039272,
      "balance_loss_clip": 1.06048942,
      "balance_loss_mlp": 1.02938366,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.2855814928221525,
      "language_loss": 0.84119129,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86359316,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 2.6050872802734375
    },
    {
      "auxiliary_loss_clip": 0.01205274,
      "auxiliary_loss_mlp": 0.0103734,
      "balance_loss_clip": 1.06206083,
      "balance_loss_mlp": 1.02732682,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.312200826647629,
      "language_loss": 0.77229249,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79471874,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.6156625747680664
    },
    {
      "auxiliary_loss_clip": 0.01201714,
      "auxiliary_loss_mlp": 0.0103334,
      "balance_loss_clip": 1.05873609,
      "balance_loss_mlp": 1.02427471,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 3.8820094870171986,
      "language_loss": 0.7092483,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.73159885,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.6687474250793457
    },
    {
      "auxiliary_loss_clip": 0.01083926,
      "auxiliary_loss_mlp": 0.01011878,
      "balance_loss_clip": 1.01968551,
      "balance_loss_mlp": 1.0098635,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7976430935275114,
      "language_loss": 0.64800495,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66896302,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.324531078338623
    },
    {
      "auxiliary_loss_clip": 0.0109611,
      "auxiliary_loss_mlp": 0.00893598,
      "balance_loss_clip": 1.02386785,
      "balance_loss_mlp": 1.00041378,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 16.785510302890245,
      "language_loss": 0.62895411,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64885122,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.1915900707244873
    },
    {
      "auxiliary_loss_clip": 0.0117205,
      "auxiliary_loss_mlp": 0.01037815,
      "balance_loss_clip": 1.05324137,
      "balance_loss_mlp": 1.02746189,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 1.942777566449906,
      "language_loss": 0.73771131,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75980997,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.7520134449005127
    },
    {
      "auxiliary_loss_clip": 0.0119963,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.05544603,
      "balance_loss_mlp": 1.02222979,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.019783483492715,
      "language_loss": 0.80829239,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.8306129,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.674581527709961
    },
    {
      "auxiliary_loss_clip": 0.01193513,
      "auxiliary_loss_mlp": 0.0103192,
      "balance_loss_clip": 1.05820978,
      "balance_loss_mlp": 1.02182317,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.983103938038123,
      "language_loss": 0.8436687,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86592305,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.655360698699951
    },
    {
      "auxiliary_loss_clip": 0.01094294,
      "auxiliary_loss_mlp": 0.01003283,
      "balance_loss_clip": 1.02285135,
      "balance_loss_mlp": 1.00126839,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.8808340668988011,
      "language_loss": 0.64196622,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66294205,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.090731382369995
    },
    {
      "auxiliary_loss_clip": 0.01204718,
      "auxiliary_loss_mlp": 0.01033186,
      "balance_loss_clip": 1.06061912,
      "balance_loss_mlp": 1.02277255,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 1.9028483127115228,
      "language_loss": 0.78744364,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80982262,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.6823904514312744
    },
    {
      "auxiliary_loss_clip": 0.01219802,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.06396878,
      "balance_loss_mlp": 1.02100587,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.405525574008905,
      "language_loss": 0.72747087,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74998558,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.6150059700012207
    },
    {
      "auxiliary_loss_clip": 0.01103936,
      "auxiliary_loss_mlp": 0.0100314,
      "balance_loss_clip": 1.0230813,
      "balance_loss_mlp": 1.00112557,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8289495678295037,
      "language_loss": 0.64875042,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.6698212,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.217871904373169
    },
    {
      "auxiliary_loss_clip": 0.01214006,
      "auxiliary_loss_mlp": 0.01031087,
      "balance_loss_clip": 1.06243944,
      "balance_loss_mlp": 1.02123427,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 1.9508111765604146,
      "language_loss": 0.87587047,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89832139,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 4.523323059082031
    },
    {
      "auxiliary_loss_clip": 0.0121413,
      "auxiliary_loss_mlp": 0.01036175,
      "balance_loss_clip": 1.06158376,
      "balance_loss_mlp": 1.02672756,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 2.4919622306101665,
      "language_loss": 0.74450117,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.76700419,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.653331995010376
    },
    {
      "auxiliary_loss_clip": 0.01204261,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.05965173,
      "balance_loss_mlp": 1.02065122,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.924138373478617,
      "language_loss": 0.77141714,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79376531,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 3.5991101264953613
    },
    {
      "auxiliary_loss_clip": 0.01218676,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 1.06136513,
      "balance_loss_mlp": 1.02353263,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.7134006569080893,
      "language_loss": 0.76778805,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79031479,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.584912061691284
    },
    {
      "auxiliary_loss_clip": 0.01163837,
      "auxiliary_loss_mlp": 0.01034443,
      "balance_loss_clip": 1.05476141,
      "balance_loss_mlp": 1.02361321,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.3709881147150154,
      "language_loss": 0.85743189,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87941474,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.737400770187378
    },
    {
      "auxiliary_loss_clip": 0.01210174,
      "auxiliary_loss_mlp": 0.01039994,
      "balance_loss_clip": 1.06089246,
      "balance_loss_mlp": 1.02909255,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 3.091362884840421,
      "language_loss": 0.91441834,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93692005,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.579332113265991
    },
    {
      "auxiliary_loss_clip": 0.01092881,
      "auxiliary_loss_mlp": 0.01002464,
      "balance_loss_clip": 1.02054036,
      "balance_loss_mlp": 1.00056839,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7760631378523402,
      "language_loss": 0.59293407,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61388755,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 4.30743145942688
    },
    {
      "auxiliary_loss_clip": 0.01188941,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 1.06071544,
      "balance_loss_mlp": 1.02335417,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 3.123194802504998,
      "language_loss": 0.80545533,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.82767999,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.7207887172698975
    },
    {
      "auxiliary_loss_clip": 0.01207628,
      "auxiliary_loss_mlp": 0.01031059,
      "balance_loss_clip": 1.05814695,
      "balance_loss_mlp": 1.02009225,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.9754918901522496,
      "language_loss": 0.83231801,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.8547048,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.7138514518737793
    },
    {
      "auxiliary_loss_clip": 0.01218225,
      "auxiliary_loss_mlp": 0.01037939,
      "balance_loss_clip": 1.06591439,
      "balance_loss_mlp": 1.02648914,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.400635434721801,
      "language_loss": 0.81138265,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83394426,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.6211721897125244
    },
    {
      "auxiliary_loss_clip": 0.01211028,
      "auxiliary_loss_mlp": 0.01034738,
      "balance_loss_clip": 1.0655539,
      "balance_loss_mlp": 1.02526677,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 2.1802875250454266,
      "language_loss": 0.81484455,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83730215,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.6543962955474854
    },
    {
      "auxiliary_loss_clip": 0.01197726,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.06054401,
      "balance_loss_mlp": 1.02341819,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 3.4454514553536724,
      "language_loss": 0.77781576,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80013132,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.6580464839935303
    },
    {
      "auxiliary_loss_clip": 0.01092928,
      "auxiliary_loss_mlp": 0.01000789,
      "balance_loss_clip": 1.02289736,
      "balance_loss_mlp": 0.99891758,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.7827310456111605,
      "language_loss": 0.54195958,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56289679,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.212181568145752
    },
    {
      "auxiliary_loss_clip": 0.01218855,
      "auxiliary_loss_mlp": 0.01038694,
      "balance_loss_clip": 1.06764436,
      "balance_loss_mlp": 1.02818584,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 1.9574811066102318,
      "language_loss": 0.77189124,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79446667,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 2.5863540172576904
    },
    {
      "auxiliary_loss_clip": 0.01193989,
      "auxiliary_loss_mlp": 0.01031787,
      "balance_loss_clip": 1.06060255,
      "balance_loss_mlp": 1.02178574,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 2.899669361320713,
      "language_loss": 0.81566346,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83792126,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 2.745176315307617
    },
    {
      "auxiliary_loss_clip": 0.01169409,
      "auxiliary_loss_mlp": 0.01035283,
      "balance_loss_clip": 1.05624866,
      "balance_loss_mlp": 1.0246973,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.795442944733594,
      "language_loss": 0.7339325,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75597936,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.8002803325653076
    },
    {
      "auxiliary_loss_clip": 0.01091494,
      "auxiliary_loss_mlp": 0.01002056,
      "balance_loss_clip": 1.0398066,
      "balance_loss_mlp": 0.9999699,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.909215300188458,
      "language_loss": 0.58247036,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60340589,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.0742979049682617
    },
    {
      "auxiliary_loss_clip": 0.01202098,
      "auxiliary_loss_mlp": 0.01036579,
      "balance_loss_clip": 1.05976009,
      "balance_loss_mlp": 1.02610087,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.843370407677242,
      "language_loss": 0.80457199,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82695872,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.863508701324463
    },
    {
      "auxiliary_loss_clip": 0.01217545,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.0653187,
      "balance_loss_mlp": 1.0224874,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.8210197244404265,
      "language_loss": 0.86983597,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89233875,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.6656737327575684
    },
    {
      "auxiliary_loss_clip": 0.01220169,
      "auxiliary_loss_mlp": 0.01036699,
      "balance_loss_clip": 1.0651859,
      "balance_loss_mlp": 1.02571404,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 2.089050284914058,
      "language_loss": 0.70746315,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73003185,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.599423408508301
    },
    {
      "auxiliary_loss_clip": 0.01194568,
      "auxiliary_loss_mlp": 0.0103522,
      "balance_loss_clip": 1.05816853,
      "balance_loss_mlp": 1.02558196,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.8602252289797974,
      "language_loss": 0.84593999,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86823797,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.8557095527648926
    },
    {
      "auxiliary_loss_clip": 0.01183354,
      "auxiliary_loss_mlp": 0.01035716,
      "balance_loss_clip": 1.05534172,
      "balance_loss_mlp": 1.02464163,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.76751754156727,
      "language_loss": 0.72378451,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.7459752,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.8079993724823
    },
    {
      "auxiliary_loss_clip": 0.0119095,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.05833435,
      "balance_loss_mlp": 1.02893305,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 2.1305809005300413,
      "language_loss": 0.84237587,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86466926,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.712794780731201
    },
    {
      "auxiliary_loss_clip": 0.01209659,
      "auxiliary_loss_mlp": 0.01034655,
      "balance_loss_clip": 1.05916357,
      "balance_loss_mlp": 1.02409351,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 2.572861053922673,
      "language_loss": 0.75191391,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77435702,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.7564923763275146
    },
    {
      "auxiliary_loss_clip": 0.01205731,
      "auxiliary_loss_mlp": 0.01032949,
      "balance_loss_clip": 1.06204748,
      "balance_loss_mlp": 1.02269697,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 3.4243940919805125,
      "language_loss": 0.77656132,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79894817,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.6347882747650146
    },
    {
      "auxiliary_loss_clip": 0.01187271,
      "auxiliary_loss_mlp": 0.01036477,
      "balance_loss_clip": 1.06000888,
      "balance_loss_mlp": 1.02581406,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.0671223054790504,
      "language_loss": 0.78623456,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80847204,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.6784579753875732
    },
    {
      "auxiliary_loss_clip": 0.0115718,
      "auxiliary_loss_mlp": 0.01043143,
      "balance_loss_clip": 1.04993105,
      "balance_loss_mlp": 1.03203332,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 2.2809743233681816,
      "language_loss": 0.76969635,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79169959,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 3.71600079536438
    },
    {
      "auxiliary_loss_clip": 0.01182219,
      "auxiliary_loss_mlp": 0.01034745,
      "balance_loss_clip": 1.05458009,
      "balance_loss_mlp": 1.02330685,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 3.2327430237857318,
      "language_loss": 0.76253635,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.784706,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 3.5929505825042725
    },
    {
      "auxiliary_loss_clip": 0.0120884,
      "auxiliary_loss_mlp": 0.01040904,
      "balance_loss_clip": 1.06376815,
      "balance_loss_mlp": 1.03074765,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.2669030095251794,
      "language_loss": 0.92566323,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94816065,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.6749541759490967
    },
    {
      "auxiliary_loss_clip": 0.01187643,
      "auxiliary_loss_mlp": 0.01034767,
      "balance_loss_clip": 1.05733633,
      "balance_loss_mlp": 1.0255636,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.6389700342055997,
      "language_loss": 0.76070857,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.78293264,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 3.5947017669677734
    },
    {
      "auxiliary_loss_clip": 0.01197413,
      "auxiliary_loss_mlp": 0.01040952,
      "balance_loss_clip": 1.0553751,
      "balance_loss_mlp": 1.03105783,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 1.9179145375213549,
      "language_loss": 0.85562485,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87800854,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.703749418258667
    },
    {
      "auxiliary_loss_clip": 0.01196694,
      "auxiliary_loss_mlp": 0.01031472,
      "balance_loss_clip": 1.06408238,
      "balance_loss_mlp": 1.0213573,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 5.876561304486342,
      "language_loss": 0.88850772,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.91078937,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.649024248123169
    },
    {
      "auxiliary_loss_clip": 0.01213254,
      "auxiliary_loss_mlp": 0.01037912,
      "balance_loss_clip": 1.06139445,
      "balance_loss_mlp": 1.02756476,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.23794577592777,
      "language_loss": 0.81416148,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.83667314,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.588191270828247
    },
    {
      "auxiliary_loss_clip": 0.0114049,
      "auxiliary_loss_mlp": 0.01043099,
      "balance_loss_clip": 1.0464462,
      "balance_loss_mlp": 1.03182137,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 2.505257134517869,
      "language_loss": 0.7159887,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73782468,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 3.8231089115142822
    },
    {
      "auxiliary_loss_clip": 0.01182523,
      "auxiliary_loss_mlp": 0.01038089,
      "balance_loss_clip": 1.05459559,
      "balance_loss_mlp": 1.0260371,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.3646526706491207,
      "language_loss": 0.89880228,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.92100841,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 2.8193109035491943
    },
    {
      "auxiliary_loss_clip": 0.01099691,
      "auxiliary_loss_mlp": 0.01007557,
      "balance_loss_clip": 1.03061378,
      "balance_loss_mlp": 1.00550675,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7883397705949732,
      "language_loss": 0.5549081,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.5759806,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.177899122238159
    },
    {
      "auxiliary_loss_clip": 0.01194084,
      "auxiliary_loss_mlp": 0.01040657,
      "balance_loss_clip": 1.05950034,
      "balance_loss_mlp": 1.02902794,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 3.4618534590911616,
      "language_loss": 0.81135499,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83370233,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.6936323642730713
    },
    {
      "auxiliary_loss_clip": 0.01173182,
      "auxiliary_loss_mlp": 0.01034041,
      "balance_loss_clip": 1.05696702,
      "balance_loss_mlp": 1.02345514,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.3149666528571453,
      "language_loss": 0.75589323,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77796549,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.7063088417053223
    },
    {
      "auxiliary_loss_clip": 0.01208863,
      "auxiliary_loss_mlp": 0.00904786,
      "balance_loss_clip": 1.06568432,
      "balance_loss_mlp": 1.00226951,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.407491595819544,
      "language_loss": 0.84761733,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86875385,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.6660666465759277
    },
    {
      "auxiliary_loss_clip": 0.01190832,
      "auxiliary_loss_mlp": 0.01040093,
      "balance_loss_clip": 1.05687594,
      "balance_loss_mlp": 1.02921546,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 2.132082249091017,
      "language_loss": 0.79304588,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81535506,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 2.672077178955078
    },
    {
      "auxiliary_loss_clip": 0.01195513,
      "auxiliary_loss_mlp": 0.01031709,
      "balance_loss_clip": 1.06245148,
      "balance_loss_mlp": 1.02158833,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.7732733321375314,
      "language_loss": 0.80725712,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82952929,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.7051334381103516
    },
    {
      "auxiliary_loss_clip": 0.01197101,
      "auxiliary_loss_mlp": 0.0103655,
      "balance_loss_clip": 1.06037939,
      "balance_loss_mlp": 1.02604818,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 2.4161664966455927,
      "language_loss": 0.81810904,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84044558,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.6631453037261963
    },
    {
      "auxiliary_loss_clip": 0.01205508,
      "auxiliary_loss_mlp": 0.01035416,
      "balance_loss_clip": 1.0631789,
      "balance_loss_mlp": 1.02496684,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 2.1604453867448528,
      "language_loss": 0.72873831,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75114751,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.7700693607330322
    },
    {
      "auxiliary_loss_clip": 0.01222589,
      "auxiliary_loss_mlp": 0.01040257,
      "balance_loss_clip": 1.06599784,
      "balance_loss_mlp": 1.02924812,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 3.0543514529118556,
      "language_loss": 0.81508899,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83771741,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.68380069732666
    },
    {
      "auxiliary_loss_clip": 0.01206344,
      "auxiliary_loss_mlp": 0.01030415,
      "balance_loss_clip": 1.059533,
      "balance_loss_mlp": 1.01949501,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.9368752805279126,
      "language_loss": 0.69958359,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.72195119,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.66838002204895
    },
    {
      "auxiliary_loss_clip": 0.01208882,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.06085157,
      "balance_loss_mlp": 1.02176428,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.6763644578584658,
      "language_loss": 0.86820608,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.89061534,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.7132229804992676
    },
    {
      "auxiliary_loss_clip": 0.01204635,
      "auxiliary_loss_mlp": 0.01037502,
      "balance_loss_clip": 1.06154895,
      "balance_loss_mlp": 1.0275538,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.757967597653913,
      "language_loss": 0.7387985,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76121992,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.924628496170044
    },
    {
      "auxiliary_loss_clip": 0.01185764,
      "auxiliary_loss_mlp": 0.01047586,
      "balance_loss_clip": 1.05657935,
      "balance_loss_mlp": 1.0360173,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 2.5213598942657343,
      "language_loss": 0.78654146,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80887502,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.6989364624023438
    },
    {
      "auxiliary_loss_clip": 0.01187165,
      "auxiliary_loss_mlp": 0.0103289,
      "balance_loss_clip": 1.057477,
      "balance_loss_mlp": 1.02206624,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 1.9667168104216541,
      "language_loss": 0.8569901,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.87919068,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.65911602973938
    },
    {
      "auxiliary_loss_clip": 0.01209053,
      "auxiliary_loss_mlp": 0.01038675,
      "balance_loss_clip": 1.06602812,
      "balance_loss_mlp": 1.02788687,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 3.5417239480709117,
      "language_loss": 0.8547188,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87719607,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.6926825046539307
    },
    {
      "auxiliary_loss_clip": 0.01193889,
      "auxiliary_loss_mlp": 0.01034072,
      "balance_loss_clip": 1.05877423,
      "balance_loss_mlp": 1.02269959,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 2.9964381837626886,
      "language_loss": 0.59876484,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62104446,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.711960792541504
    },
    {
      "auxiliary_loss_clip": 0.01184386,
      "auxiliary_loss_mlp": 0.01032773,
      "balance_loss_clip": 1.05706632,
      "balance_loss_mlp": 1.02153122,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.1734995040039737,
      "language_loss": 0.9273814,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94955301,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.735488176345825
    },
    {
      "auxiliary_loss_clip": 0.0120487,
      "auxiliary_loss_mlp": 0.01045119,
      "balance_loss_clip": 1.06149936,
      "balance_loss_mlp": 1.03312683,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.202648197165798,
      "language_loss": 0.72006929,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.74256921,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 4.010451793670654
    },
    {
      "auxiliary_loss_clip": 0.01188364,
      "auxiliary_loss_mlp": 0.01036625,
      "balance_loss_clip": 1.0589366,
      "balance_loss_mlp": 1.02621794,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.2428938901855835,
      "language_loss": 0.8301034,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85235327,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.6666488647460938
    },
    {
      "auxiliary_loss_clip": 0.01188545,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 1.0599153,
      "balance_loss_mlp": 1.02833533,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.8018439455127908,
      "language_loss": 0.73936033,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76163191,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 3.7137763500213623
    },
    {
      "auxiliary_loss_clip": 0.0119194,
      "auxiliary_loss_mlp": 0.010347,
      "balance_loss_clip": 1.0612222,
      "balance_loss_mlp": 1.0245254,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 2.437728433608776,
      "language_loss": 0.86255676,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.88482314,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 3.7401673793792725
    },
    {
      "auxiliary_loss_clip": 0.01215979,
      "auxiliary_loss_mlp": 0.01029137,
      "balance_loss_clip": 1.06327653,
      "balance_loss_mlp": 1.01903367,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 2.0282990468798574,
      "language_loss": 0.92686766,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94931877,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.6060359477996826
    },
    {
      "auxiliary_loss_clip": 0.01206162,
      "auxiliary_loss_mlp": 0.01033173,
      "balance_loss_clip": 1.06388402,
      "balance_loss_mlp": 1.02137744,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 5.011843758164802,
      "language_loss": 0.73629093,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75868428,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.8660430908203125
    },
    {
      "auxiliary_loss_clip": 0.01197627,
      "auxiliary_loss_mlp": 0.01032546,
      "balance_loss_clip": 1.0618602,
      "balance_loss_mlp": 1.0225203,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.7654766367629215,
      "language_loss": 0.88422728,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90652907,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.6386778354644775
    },
    {
      "auxiliary_loss_clip": 0.0118509,
      "auxiliary_loss_mlp": 0.01038516,
      "balance_loss_clip": 1.06231463,
      "balance_loss_mlp": 1.02865124,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 2.037019226484461,
      "language_loss": 0.76506031,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78729641,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.699355363845825
    },
    {
      "auxiliary_loss_clip": 0.01217974,
      "auxiliary_loss_mlp": 0.00904697,
      "balance_loss_clip": 1.06681275,
      "balance_loss_mlp": 1.00231791,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.6113725407444623,
      "language_loss": 0.81006485,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.83129162,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 3.590440273284912
    },
    {
      "auxiliary_loss_clip": 0.01195842,
      "auxiliary_loss_mlp": 0.01036126,
      "balance_loss_clip": 1.05961585,
      "balance_loss_mlp": 1.02510524,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.7967789861384413,
      "language_loss": 0.70246977,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.7247895,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.650942325592041
    },
    {
      "auxiliary_loss_clip": 0.01192246,
      "auxiliary_loss_mlp": 0.01038957,
      "balance_loss_clip": 1.05938411,
      "balance_loss_mlp": 1.02720892,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 2.8518633180209187,
      "language_loss": 0.81699049,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.83930254,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.771662712097168
    },
    {
      "auxiliary_loss_clip": 0.01209646,
      "auxiliary_loss_mlp": 0.01035828,
      "balance_loss_clip": 1.06053162,
      "balance_loss_mlp": 1.02521205,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 2.044973664739669,
      "language_loss": 0.79806674,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.82052147,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.625764846801758
    },
    {
      "auxiliary_loss_clip": 0.01206696,
      "auxiliary_loss_mlp": 0.01033387,
      "balance_loss_clip": 1.06202674,
      "balance_loss_mlp": 1.02255058,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.7162498816224003,
      "language_loss": 0.74824607,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.77064687,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.6971395015716553
    },
    {
      "auxiliary_loss_clip": 0.01216444,
      "auxiliary_loss_mlp": 0.0103595,
      "balance_loss_clip": 1.06532693,
      "balance_loss_mlp": 1.02659202,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.6144942276539564,
      "language_loss": 0.80578995,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82831383,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 2.536987066268921
    },
    {
      "auxiliary_loss_clip": 0.01101405,
      "auxiliary_loss_mlp": 0.01013447,
      "balance_loss_clip": 1.03047585,
      "balance_loss_mlp": 1.01120627,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8354661575329285,
      "language_loss": 0.577986,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59913445,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.30566143989563
    },
    {
      "auxiliary_loss_clip": 0.01188018,
      "auxiliary_loss_mlp": 0.0103103,
      "balance_loss_clip": 1.05979633,
      "balance_loss_mlp": 1.02122509,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 3.1000160729220516,
      "language_loss": 0.67423058,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69642103,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 2.7035329341888428
    },
    {
      "auxiliary_loss_clip": 0.01218003,
      "auxiliary_loss_mlp": 0.01038444,
      "balance_loss_clip": 1.06613612,
      "balance_loss_mlp": 1.02907443,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 4.189404085198612,
      "language_loss": 0.74243343,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.76499796,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.634143352508545
    },
    {
      "auxiliary_loss_clip": 0.01221913,
      "auxiliary_loss_mlp": 0.01034861,
      "balance_loss_clip": 1.06724477,
      "balance_loss_mlp": 1.0237751,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 2.019682874401987,
      "language_loss": 0.73289835,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.7554661,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.6751956939697266
    },
    {
      "auxiliary_loss_clip": 0.01113414,
      "auxiliary_loss_mlp": 0.01009191,
      "balance_loss_clip": 1.04183662,
      "balance_loss_mlp": 1.00689054,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8255605806038673,
      "language_loss": 0.58191991,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60314596,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.216853141784668
    },
    {
      "auxiliary_loss_clip": 0.01212157,
      "auxiliary_loss_mlp": 0.01035251,
      "balance_loss_clip": 1.06072283,
      "balance_loss_mlp": 1.02479076,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 2.6518750963671085,
      "language_loss": 0.85957527,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.88204932,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.647081136703491
    },
    {
      "auxiliary_loss_clip": 0.01185808,
      "auxiliary_loss_mlp": 0.01036262,
      "balance_loss_clip": 1.05763006,
      "balance_loss_mlp": 1.02563405,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.1557168539666316,
      "language_loss": 0.76479304,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78701377,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.800436496734619
    },
    {
      "auxiliary_loss_clip": 0.01215243,
      "auxiliary_loss_mlp": 0.01034656,
      "balance_loss_clip": 1.06364918,
      "balance_loss_mlp": 1.02418303,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.83938473107612,
      "language_loss": 0.77129591,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79379487,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.698094606399536
    },
    {
      "auxiliary_loss_clip": 0.01210763,
      "auxiliary_loss_mlp": 0.00905162,
      "balance_loss_clip": 1.05940342,
      "balance_loss_mlp": 1.0022608,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.01567237879997,
      "language_loss": 0.82173574,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.84289497,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.658845901489258
    },
    {
      "auxiliary_loss_clip": 0.01185207,
      "auxiliary_loss_mlp": 0.01038427,
      "balance_loss_clip": 1.05444813,
      "balance_loss_mlp": 1.02784705,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.2639543692643946,
      "language_loss": 0.76529056,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.78752697,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.852957248687744
    },
    {
      "auxiliary_loss_clip": 0.01215329,
      "auxiliary_loss_mlp": 0.01034365,
      "balance_loss_clip": 1.06219578,
      "balance_loss_mlp": 1.02478635,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.7220796730991046,
      "language_loss": 0.70105594,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72355288,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.5759081840515137
    },
    {
      "auxiliary_loss_clip": 0.01201487,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.06128681,
      "balance_loss_mlp": 1.02180004,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.2356762615754646,
      "language_loss": 0.89079571,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91313195,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.603527307510376
    },
    {
      "auxiliary_loss_clip": 0.01191393,
      "auxiliary_loss_mlp": 0.01031583,
      "balance_loss_clip": 1.0606252,
      "balance_loss_mlp": 1.02282119,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.572623831498076,
      "language_loss": 0.76518595,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.78741574,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.742393970489502
    },
    {
      "auxiliary_loss_clip": 0.01169944,
      "auxiliary_loss_mlp": 0.01035744,
      "balance_loss_clip": 1.05420017,
      "balance_loss_mlp": 1.02550352,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 2.001337135453204,
      "language_loss": 0.87020159,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89225847,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 3.7495367527008057
    },
    {
      "auxiliary_loss_clip": 0.01188783,
      "auxiliary_loss_mlp": 0.01031144,
      "balance_loss_clip": 1.05853677,
      "balance_loss_mlp": 1.02095199,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.194834798650995,
      "language_loss": 0.78846025,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.81065953,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 3.6308138370513916
    },
    {
      "auxiliary_loss_clip": 0.01194656,
      "auxiliary_loss_mlp": 0.0090398,
      "balance_loss_clip": 1.06202388,
      "balance_loss_mlp": 1.00220346,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.2285499014190138,
      "language_loss": 0.89695573,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.91794211,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.623140573501587
    },
    {
      "auxiliary_loss_clip": 0.01205872,
      "auxiliary_loss_mlp": 0.01031455,
      "balance_loss_clip": 1.05995417,
      "balance_loss_mlp": 1.02132201,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.4137317912824,
      "language_loss": 0.78572041,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.80809367,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 3.501887321472168
    },
    {
      "auxiliary_loss_clip": 0.01194829,
      "auxiliary_loss_mlp": 0.01036566,
      "balance_loss_clip": 1.06027877,
      "balance_loss_mlp": 1.02585542,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 1.9011608844021473,
      "language_loss": 0.71045071,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.7327646,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.67968487739563
    },
    {
      "auxiliary_loss_clip": 0.01213398,
      "auxiliary_loss_mlp": 0.01034308,
      "balance_loss_clip": 1.06195176,
      "balance_loss_mlp": 1.02527189,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.1581178601257425,
      "language_loss": 0.80978304,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83226007,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.578207492828369
    },
    {
      "auxiliary_loss_clip": 0.01177702,
      "auxiliary_loss_mlp": 0.00904833,
      "balance_loss_clip": 1.05720544,
      "balance_loss_mlp": 1.00255585,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.9350826826023735,
      "language_loss": 0.83692914,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.85775447,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.687448740005493
    },
    {
      "auxiliary_loss_clip": 0.01206066,
      "auxiliary_loss_mlp": 0.01041502,
      "balance_loss_clip": 1.05831528,
      "balance_loss_mlp": 1.03052902,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 2.39403806340173,
      "language_loss": 0.68620443,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.70868003,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 3.595782518386841
    },
    {
      "auxiliary_loss_clip": 0.01217231,
      "auxiliary_loss_mlp": 0.01038737,
      "balance_loss_clip": 1.06472039,
      "balance_loss_mlp": 1.0282824,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 2.706227472370208,
      "language_loss": 0.6726898,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.69524944,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.677196741104126
    },
    {
      "auxiliary_loss_clip": 0.01177921,
      "auxiliary_loss_mlp": 0.01033804,
      "balance_loss_clip": 1.058689,
      "balance_loss_mlp": 1.02371287,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 1.893918591331241,
      "language_loss": 0.85886949,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88098675,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.8749988079071045
    },
    {
      "auxiliary_loss_clip": 0.01189621,
      "auxiliary_loss_mlp": 0.0103514,
      "balance_loss_clip": 1.0565207,
      "balance_loss_mlp": 1.02429819,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.2129225441017626,
      "language_loss": 0.7937094,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81595695,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.747678756713867
    },
    {
      "auxiliary_loss_clip": 0.01188215,
      "auxiliary_loss_mlp": 0.01035342,
      "balance_loss_clip": 1.05799294,
      "balance_loss_mlp": 1.02562666,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.7268421757383994,
      "language_loss": 0.82203496,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84427059,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.6284658908843994
    },
    {
      "auxiliary_loss_clip": 0.01199088,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.06043315,
      "balance_loss_mlp": 1.02229738,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.782202874496123,
      "language_loss": 0.69328928,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71560144,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.683263063430786
    },
    {
      "auxiliary_loss_clip": 0.01168662,
      "auxiliary_loss_mlp": 0.01031143,
      "balance_loss_clip": 1.05452263,
      "balance_loss_mlp": 1.02090311,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 3.382450749475289,
      "language_loss": 0.74128139,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76327938,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.7833356857299805
    },
    {
      "auxiliary_loss_clip": 0.01178356,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.05373693,
      "balance_loss_mlp": 1.02318692,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 2.311933305008953,
      "language_loss": 0.80462253,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.82673454,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 2.708843231201172
    },
    {
      "auxiliary_loss_clip": 0.01101334,
      "auxiliary_loss_mlp": 0.010024,
      "balance_loss_clip": 1.02678967,
      "balance_loss_mlp": 1.00020647,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8453771483832859,
      "language_loss": 0.57906353,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.60010087,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.1955208778381348
    },
    {
      "auxiliary_loss_clip": 0.01195056,
      "auxiliary_loss_mlp": 0.01035208,
      "balance_loss_clip": 1.05547655,
      "balance_loss_mlp": 1.02473545,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.236251207463099,
      "language_loss": 0.83353925,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85584182,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.692971706390381
    },
    {
      "auxiliary_loss_clip": 0.01183372,
      "auxiliary_loss_mlp": 0.01034055,
      "balance_loss_clip": 1.05782461,
      "balance_loss_mlp": 1.02508438,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 2.198581651144236,
      "language_loss": 0.80469108,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82686538,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.73940372467041
    },
    {
      "auxiliary_loss_clip": 0.01194994,
      "auxiliary_loss_mlp": 0.01038944,
      "balance_loss_clip": 1.05960858,
      "balance_loss_mlp": 1.02968705,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 5.421778046694335,
      "language_loss": 0.77883601,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80117542,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.62760853767395
    },
    {
      "auxiliary_loss_clip": 0.01185295,
      "auxiliary_loss_mlp": 0.01032711,
      "balance_loss_clip": 1.05625701,
      "balance_loss_mlp": 1.02286434,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.5567261810896293,
      "language_loss": 0.79701209,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.81919211,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.7336525917053223
    },
    {
      "auxiliary_loss_clip": 0.01216529,
      "auxiliary_loss_mlp": 0.0103652,
      "balance_loss_clip": 1.06374621,
      "balance_loss_mlp": 1.02685833,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 2.8551231075291947,
      "language_loss": 0.86411715,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.8866477,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.6758594512939453
    },
    {
      "auxiliary_loss_clip": 0.01192595,
      "auxiliary_loss_mlp": 0.00904551,
      "balance_loss_clip": 1.05944705,
      "balance_loss_mlp": 1.00248551,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 3.0200810463975247,
      "language_loss": 0.89395642,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91492784,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.867889165878296
    },
    {
      "auxiliary_loss_clip": 0.01185717,
      "auxiliary_loss_mlp": 0.01033197,
      "balance_loss_clip": 1.0579896,
      "balance_loss_mlp": 1.02274776,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.3568317155710177,
      "language_loss": 0.90356696,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.9257561,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.7238094806671143
    },
    {
      "auxiliary_loss_clip": 0.011814,
      "auxiliary_loss_mlp": 0.01042896,
      "balance_loss_clip": 1.05974519,
      "balance_loss_mlp": 1.03146958,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 3.9911178588308514,
      "language_loss": 0.87151504,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.89375794,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.6660611629486084
    },
    {
      "auxiliary_loss_clip": 0.0119933,
      "auxiliary_loss_mlp": 0.01044974,
      "balance_loss_clip": 1.06217515,
      "balance_loss_mlp": 1.03341615,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 3.379972518189515,
      "language_loss": 0.8974427,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91988575,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.6902732849121094
    },
    {
      "auxiliary_loss_clip": 0.01205709,
      "auxiliary_loss_mlp": 0.01038534,
      "balance_loss_clip": 1.05970216,
      "balance_loss_mlp": 1.02890825,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.4751255318766985,
      "language_loss": 0.79111636,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81355882,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.6603236198425293
    },
    {
      "auxiliary_loss_clip": 0.01181754,
      "auxiliary_loss_mlp": 0.01043623,
      "balance_loss_clip": 1.05475616,
      "balance_loss_mlp": 1.03265548,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 3.1107427389318114,
      "language_loss": 0.74395967,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.76621342,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 3.7100021839141846
    },
    {
      "auxiliary_loss_clip": 0.01207338,
      "auxiliary_loss_mlp": 0.01042281,
      "balance_loss_clip": 1.06088591,
      "balance_loss_mlp": 1.03152847,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 4.050961024889294,
      "language_loss": 0.84596455,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86846071,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.6369857788085938
    },
    {
      "auxiliary_loss_clip": 0.0120193,
      "auxiliary_loss_mlp": 0.01042468,
      "balance_loss_clip": 1.05989456,
      "balance_loss_mlp": 1.03222835,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.2435773905026744,
      "language_loss": 0.83538818,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85783213,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 3.5589711666107178
    },
    {
      "auxiliary_loss_clip": 0.01216102,
      "auxiliary_loss_mlp": 0.01042989,
      "balance_loss_clip": 1.06240284,
      "balance_loss_mlp": 1.03181338,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.4687360483743337,
      "language_loss": 0.85584754,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87843847,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 3.5316076278686523
    },
    {
      "auxiliary_loss_clip": 0.01216561,
      "auxiliary_loss_mlp": 0.01033234,
      "balance_loss_clip": 1.06386137,
      "balance_loss_mlp": 1.02332163,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.8908778719596373,
      "language_loss": 0.75075865,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.7732566,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.653426170349121
    },
    {
      "auxiliary_loss_clip": 0.01206901,
      "auxiliary_loss_mlp": 0.01039076,
      "balance_loss_clip": 1.06467819,
      "balance_loss_mlp": 1.02940214,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.9319760852731123,
      "language_loss": 0.72207057,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74453032,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.627694845199585
    },
    {
      "auxiliary_loss_clip": 0.01165723,
      "auxiliary_loss_mlp": 0.01039377,
      "balance_loss_clip": 1.05550325,
      "balance_loss_mlp": 1.02905393,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 2.3922584939602225,
      "language_loss": 0.74416518,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.76621616,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.972104787826538
    },
    {
      "auxiliary_loss_clip": 0.01220362,
      "auxiliary_loss_mlp": 0.00904998,
      "balance_loss_clip": 1.06403589,
      "balance_loss_mlp": 1.00257432,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.7733303701706526,
      "language_loss": 0.87762547,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89887905,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.755472183227539
    },
    {
      "auxiliary_loss_clip": 0.01151771,
      "auxiliary_loss_mlp": 0.01034639,
      "balance_loss_clip": 1.04979467,
      "balance_loss_mlp": 1.02443516,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.9711027705958457,
      "language_loss": 0.88277602,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90464014,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 3.89585542678833
    },
    {
      "auxiliary_loss_clip": 0.01214415,
      "auxiliary_loss_mlp": 0.01038749,
      "balance_loss_clip": 1.06357408,
      "balance_loss_mlp": 1.02945709,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.7928557493888555,
      "language_loss": 0.78169954,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80423117,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.620002269744873
    },
    {
      "auxiliary_loss_clip": 0.01206385,
      "auxiliary_loss_mlp": 0.01039936,
      "balance_loss_clip": 1.06249237,
      "balance_loss_mlp": 1.02937961,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 2.3198088346209382,
      "language_loss": 0.91272885,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93519211,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.70620059967041
    },
    {
      "auxiliary_loss_clip": 0.01204526,
      "auxiliary_loss_mlp": 0.01038399,
      "balance_loss_clip": 1.0594027,
      "balance_loss_mlp": 1.02865958,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 3.2754461979006164,
      "language_loss": 0.7657153,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.78814459,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.666642189025879
    },
    {
      "auxiliary_loss_clip": 0.01106993,
      "auxiliary_loss_mlp": 0.01006966,
      "balance_loss_clip": 1.03219366,
      "balance_loss_mlp": 1.00491512,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7441929762180449,
      "language_loss": 0.57898164,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.60012126,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.3350512981414795
    },
    {
      "auxiliary_loss_clip": 0.01201605,
      "auxiliary_loss_mlp": 0.010391,
      "balance_loss_clip": 1.06533003,
      "balance_loss_mlp": 1.02943254,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.2760531423585406,
      "language_loss": 0.73763621,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76004326,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 2.852180004119873
    },
    {
      "auxiliary_loss_clip": 0.01207706,
      "auxiliary_loss_mlp": 0.01039132,
      "balance_loss_clip": 1.0634563,
      "balance_loss_mlp": 1.03011394,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 2.36956323943769,
      "language_loss": 0.72572458,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74819291,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 2.6652708053588867
    },
    {
      "auxiliary_loss_clip": 0.01217612,
      "auxiliary_loss_mlp": 0.01037473,
      "balance_loss_clip": 1.06503987,
      "balance_loss_mlp": 1.02756679,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.0768945659587805,
      "language_loss": 0.74139071,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76394153,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.6718409061431885
    },
    {
      "auxiliary_loss_clip": 0.01180713,
      "auxiliary_loss_mlp": 0.01041847,
      "balance_loss_clip": 1.06144118,
      "balance_loss_mlp": 1.03202367,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.1457949395498277,
      "language_loss": 0.82524318,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84746873,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.7453136444091797
    },
    {
      "auxiliary_loss_clip": 0.01159297,
      "auxiliary_loss_mlp": 0.01031547,
      "balance_loss_clip": 1.04991937,
      "balance_loss_mlp": 1.02197456,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.7309299904900628,
      "language_loss": 0.86429322,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88620174,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.8179163932800293
    },
    {
      "auxiliary_loss_clip": 0.01192727,
      "auxiliary_loss_mlp": 0.01035595,
      "balance_loss_clip": 1.06175232,
      "balance_loss_mlp": 1.02496195,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.634495709960651,
      "language_loss": 0.77407527,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79635847,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.8305397033691406
    },
    {
      "auxiliary_loss_clip": 0.01179843,
      "auxiliary_loss_mlp": 0.01035961,
      "balance_loss_clip": 1.0554179,
      "balance_loss_mlp": 1.02457082,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 3.09658048892162,
      "language_loss": 0.83120298,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.85336107,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.7535369396209717
    },
    {
      "auxiliary_loss_clip": 0.01115551,
      "auxiliary_loss_mlp": 0.01001283,
      "balance_loss_clip": 1.0305407,
      "balance_loss_mlp": 0.9996382,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.8038053059004147,
      "language_loss": 0.55001581,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.5711841,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.184544086456299
    },
    {
      "auxiliary_loss_clip": 0.01197139,
      "auxiliary_loss_mlp": 0.01032952,
      "balance_loss_clip": 1.06059504,
      "balance_loss_mlp": 1.02274191,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 4.985886043169811,
      "language_loss": 0.72909641,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.75139731,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.647899627685547
    },
    {
      "auxiliary_loss_clip": 0.01188825,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.061625,
      "balance_loss_mlp": 1.02251768,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.2669673915126856,
      "language_loss": 0.85318571,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87539256,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.7121741771698
    },
    {
      "auxiliary_loss_clip": 0.01167524,
      "auxiliary_loss_mlp": 0.01033319,
      "balance_loss_clip": 1.05446637,
      "balance_loss_mlp": 1.02235126,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 11.728755866765425,
      "language_loss": 0.72497058,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.746979,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.732081174850464
    },
    {
      "auxiliary_loss_clip": 0.01190908,
      "auxiliary_loss_mlp": 0.01034646,
      "balance_loss_clip": 1.05657232,
      "balance_loss_mlp": 1.02472818,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 2.424949611064564,
      "language_loss": 0.82824552,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.85050118,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.7690765857696533
    },
    {
      "auxiliary_loss_clip": 0.01113791,
      "auxiliary_loss_mlp": 0.00893835,
      "balance_loss_clip": 1.02947974,
      "balance_loss_mlp": 1.001158,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.792747247144604,
      "language_loss": 0.64323163,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66330791,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.0967540740966797
    },
    {
      "auxiliary_loss_clip": 0.0120046,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.06028914,
      "balance_loss_mlp": 1.02408004,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.258337472361947,
      "language_loss": 0.89765155,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91998982,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.7212626934051514
    },
    {
      "auxiliary_loss_clip": 0.01205071,
      "auxiliary_loss_mlp": 0.01037502,
      "balance_loss_clip": 1.06167448,
      "balance_loss_mlp": 1.02764952,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 1.6799255158217787,
      "language_loss": 0.81707585,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.83950162,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 3.646104335784912
    },
    {
      "auxiliary_loss_clip": 0.01191656,
      "auxiliary_loss_mlp": 0.01045288,
      "balance_loss_clip": 1.05669224,
      "balance_loss_mlp": 1.03501225,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 1.9231191986017007,
      "language_loss": 0.86001587,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88238537,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 3.5976510047912598
    },
    {
      "auxiliary_loss_clip": 0.01213878,
      "auxiliary_loss_mlp": 0.01044913,
      "balance_loss_clip": 1.06384349,
      "balance_loss_mlp": 1.0345242,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.9719580574943911,
      "language_loss": 0.77539301,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.7979809,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.719351291656494
    },
    {
      "auxiliary_loss_clip": 0.01212684,
      "auxiliary_loss_mlp": 0.01033496,
      "balance_loss_clip": 1.06302691,
      "balance_loss_mlp": 1.0239532,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 3.7977950804009586,
      "language_loss": 0.7604394,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78290129,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 3.5136914253234863
    },
    {
      "auxiliary_loss_clip": 0.01195054,
      "auxiliary_loss_mlp": 0.01041715,
      "balance_loss_clip": 1.06119835,
      "balance_loss_mlp": 1.03117752,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 1.7790375154156841,
      "language_loss": 0.75476551,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77713323,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.792060375213623
    },
    {
      "auxiliary_loss_clip": 0.01191067,
      "auxiliary_loss_mlp": 0.01031396,
      "balance_loss_clip": 1.05932486,
      "balance_loss_mlp": 1.02109051,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.643954979625539,
      "language_loss": 0.82050705,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84273165,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.79944109916687
    },
    {
      "auxiliary_loss_clip": 0.0119899,
      "auxiliary_loss_mlp": 0.0103076,
      "balance_loss_clip": 1.05690193,
      "balance_loss_mlp": 1.0208118,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.2138734193003855,
      "language_loss": 0.81449485,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.83679235,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.5447490215301514
    },
    {
      "auxiliary_loss_clip": 0.01187108,
      "auxiliary_loss_mlp": 0.01043047,
      "balance_loss_clip": 1.05775881,
      "balance_loss_mlp": 1.03286052,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 1.7415161599067863,
      "language_loss": 0.76169467,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78399628,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 3.629297971725464
    },
    {
      "auxiliary_loss_clip": 0.01204689,
      "auxiliary_loss_mlp": 0.01032647,
      "balance_loss_clip": 1.06188703,
      "balance_loss_mlp": 1.0229975,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 2.0530188512361196,
      "language_loss": 0.81052709,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.8329004,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.635937452316284
    },
    {
      "auxiliary_loss_clip": 0.01165948,
      "auxiliary_loss_mlp": 0.0090321,
      "balance_loss_clip": 1.05616212,
      "balance_loss_mlp": 1.00245535,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.8053028638963484,
      "language_loss": 0.89856422,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.91925579,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 2.772287368774414
    },
    {
      "auxiliary_loss_clip": 0.01192097,
      "auxiliary_loss_mlp": 0.01033195,
      "balance_loss_clip": 1.05965233,
      "balance_loss_mlp": 1.02349174,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 2.595512710248464,
      "language_loss": 0.80175233,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82400531,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.7612879276275635
    },
    {
      "auxiliary_loss_clip": 0.01212925,
      "auxiliary_loss_mlp": 0.01034875,
      "balance_loss_clip": 1.06273413,
      "balance_loss_mlp": 1.025177,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.8351103272652396,
      "language_loss": 0.82699728,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.84947526,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.7368812561035156
    },
    {
      "auxiliary_loss_clip": 0.01091115,
      "auxiliary_loss_mlp": 0.01007307,
      "balance_loss_clip": 1.02278996,
      "balance_loss_mlp": 1.00550735,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7362481567962061,
      "language_loss": 0.61969423,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64067841,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.279331922531128
    },
    {
      "auxiliary_loss_clip": 0.01182571,
      "auxiliary_loss_mlp": 0.01037299,
      "balance_loss_clip": 1.05739379,
      "balance_loss_mlp": 1.02770865,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 2.192810139428342,
      "language_loss": 0.75480306,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.7770018,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.831547975540161
    },
    {
      "auxiliary_loss_clip": 0.0120687,
      "auxiliary_loss_mlp": 0.01039364,
      "balance_loss_clip": 1.0597353,
      "balance_loss_mlp": 1.02935696,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.826637233025821,
      "language_loss": 0.82687223,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.84933454,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.632439136505127
    },
    {
      "auxiliary_loss_clip": 0.0120807,
      "auxiliary_loss_mlp": 0.01039783,
      "balance_loss_clip": 1.06095755,
      "balance_loss_mlp": 1.03038299,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 2.607548859259959,
      "language_loss": 0.89777929,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.92025781,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.627028465270996
    },
    {
      "auxiliary_loss_clip": 0.01206307,
      "auxiliary_loss_mlp": 0.01037554,
      "balance_loss_clip": 1.05922508,
      "balance_loss_mlp": 1.0272069,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 4.176822993299808,
      "language_loss": 0.73563623,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75807482,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.658674478530884
    },
    {
      "auxiliary_loss_clip": 0.01161761,
      "auxiliary_loss_mlp": 0.0104197,
      "balance_loss_clip": 1.04730749,
      "balance_loss_mlp": 1.03280294,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.040822708148546,
      "language_loss": 0.74972332,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77176058,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.7098400592803955
    },
    {
      "auxiliary_loss_clip": 0.0120802,
      "auxiliary_loss_mlp": 0.01034096,
      "balance_loss_clip": 1.0597415,
      "balance_loss_mlp": 1.02428484,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 2.6901913103329806,
      "language_loss": 0.8077932,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.83021438,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.6400580406188965
    },
    {
      "auxiliary_loss_clip": 0.0120626,
      "auxiliary_loss_mlp": 0.01034774,
      "balance_loss_clip": 1.06109881,
      "balance_loss_mlp": 1.02549386,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 5.222060924316616,
      "language_loss": 0.87459993,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89701021,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.623643159866333
    },
    {
      "auxiliary_loss_clip": 0.01190713,
      "auxiliary_loss_mlp": 0.010354,
      "balance_loss_clip": 1.06128299,
      "balance_loss_mlp": 1.02638245,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 2.0023982224046377,
      "language_loss": 0.80055094,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82281208,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.6401262283325195
    },
    {
      "auxiliary_loss_clip": 0.01185766,
      "auxiliary_loss_mlp": 0.01039357,
      "balance_loss_clip": 1.05940914,
      "balance_loss_mlp": 1.02988029,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.8788099260766957,
      "language_loss": 0.79223543,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81448662,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.713852882385254
    },
    {
      "auxiliary_loss_clip": 0.01197239,
      "auxiliary_loss_mlp": 0.00904004,
      "balance_loss_clip": 1.06093812,
      "balance_loss_mlp": 1.00254095,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.6094035879006734,
      "language_loss": 0.74778587,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.76879835,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.6521828174591064
    },
    {
      "auxiliary_loss_clip": 0.01203409,
      "auxiliary_loss_mlp": 0.01036807,
      "balance_loss_clip": 1.05984831,
      "balance_loss_mlp": 1.02687097,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 2.375111982768579,
      "language_loss": 0.79956198,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82196414,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.7214081287384033
    },
    {
      "auxiliary_loss_clip": 0.01170906,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.05608702,
      "balance_loss_mlp": 1.02036476,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.6810386074322956,
      "language_loss": 0.64491689,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66691905,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.7343854904174805
    },
    {
      "auxiliary_loss_clip": 0.01214684,
      "auxiliary_loss_mlp": 0.00903805,
      "balance_loss_clip": 1.06574047,
      "balance_loss_mlp": 1.00274527,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.7605838182624878,
      "language_loss": 0.82762325,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.84880811,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.681864023208618
    },
    {
      "auxiliary_loss_clip": 0.01200317,
      "auxiliary_loss_mlp": 0.01038766,
      "balance_loss_clip": 1.06120348,
      "balance_loss_mlp": 1.02951539,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 1.8947814325770862,
      "language_loss": 0.80571681,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82810766,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.6223437786102295
    },
    {
      "auxiliary_loss_clip": 0.0117252,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.0501796,
      "balance_loss_mlp": 1.0203675,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.905233958591938,
      "language_loss": 0.66589963,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.68791854,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 3.6010329723358154
    },
    {
      "auxiliary_loss_clip": 0.01110457,
      "auxiliary_loss_mlp": 0.01002396,
      "balance_loss_clip": 1.03576255,
      "balance_loss_mlp": 1.00053596,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.7942617386778168,
      "language_loss": 0.61991668,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64104521,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 4.1470935344696045
    },
    {
      "auxiliary_loss_clip": 0.01188745,
      "auxiliary_loss_mlp": 0.01033515,
      "balance_loss_clip": 1.06066036,
      "balance_loss_mlp": 1.02457452,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.950489146935389,
      "language_loss": 0.74243945,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76466203,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 3.771650552749634
    },
    {
      "auxiliary_loss_clip": 0.01198244,
      "auxiliary_loss_mlp": 0.01030428,
      "balance_loss_clip": 1.05959773,
      "balance_loss_mlp": 1.02156496,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.692082201276919,
      "language_loss": 0.69565493,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71794164,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.656320095062256
    },
    {
      "auxiliary_loss_clip": 0.01186569,
      "auxiliary_loss_mlp": 0.01031332,
      "balance_loss_clip": 1.05974627,
      "balance_loss_mlp": 1.02121675,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 6.045930938586618,
      "language_loss": 0.8325516,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.85473061,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.6836447715759277
    },
    {
      "auxiliary_loss_clip": 0.01195868,
      "auxiliary_loss_mlp": 0.01033938,
      "balance_loss_clip": 1.06200361,
      "balance_loss_mlp": 1.02518773,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 1.9860665441535645,
      "language_loss": 0.73271888,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75501692,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.615532875061035
    },
    {
      "auxiliary_loss_clip": 0.01182379,
      "auxiliary_loss_mlp": 0.00903476,
      "balance_loss_clip": 1.05523038,
      "balance_loss_mlp": 1.00257683,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.084708982327609,
      "language_loss": 0.67949617,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.70035481,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.7227509021759033
    },
    {
      "auxiliary_loss_clip": 0.01149367,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.04741001,
      "balance_loss_mlp": 1.03072965,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.8130770701253216,
      "language_loss": 0.80844331,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.83033609,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.7350828647613525
    },
    {
      "auxiliary_loss_clip": 0.01202214,
      "auxiliary_loss_mlp": 0.01035016,
      "balance_loss_clip": 1.06295037,
      "balance_loss_mlp": 1.02519357,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 4.553034569319415,
      "language_loss": 0.69819462,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.72056693,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 2.6075327396392822
    },
    {
      "auxiliary_loss_clip": 0.0120067,
      "auxiliary_loss_mlp": 0.01035488,
      "balance_loss_clip": 1.06038618,
      "balance_loss_mlp": 1.02595735,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 5.165754067403642,
      "language_loss": 0.73871833,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76107991,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.6420021057128906
    },
    {
      "auxiliary_loss_clip": 0.01209132,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.06005549,
      "balance_loss_mlp": 1.01933503,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.389728995170498,
      "language_loss": 0.77966022,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80204117,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.5958774089813232
    },
    {
      "auxiliary_loss_clip": 0.01211599,
      "auxiliary_loss_mlp": 0.01031482,
      "balance_loss_clip": 1.06288886,
      "balance_loss_mlp": 1.02267289,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.0251865561622093,
      "language_loss": 0.83323377,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85566461,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.6375393867492676
    },
    {
      "auxiliary_loss_clip": 0.01184537,
      "auxiliary_loss_mlp": 0.01029061,
      "balance_loss_clip": 1.05191135,
      "balance_loss_mlp": 1.020818,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.725281618819447,
      "language_loss": 0.74981999,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77195603,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.6326000690460205
    },
    {
      "auxiliary_loss_clip": 0.01210956,
      "auxiliary_loss_mlp": 0.01030404,
      "balance_loss_clip": 1.06148481,
      "balance_loss_mlp": 1.02101612,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 2.222439833298478,
      "language_loss": 0.80566478,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82807839,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.7049310207366943
    },
    {
      "auxiliary_loss_clip": 0.01151508,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.04781437,
      "balance_loss_mlp": 1.02316439,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.5658268844559413,
      "language_loss": 0.67891508,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.70076501,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.8330070972442627
    },
    {
      "auxiliary_loss_clip": 0.01201107,
      "auxiliary_loss_mlp": 0.01032564,
      "balance_loss_clip": 1.06167305,
      "balance_loss_mlp": 1.02367711,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 2.0447242510285006,
      "language_loss": 0.83924878,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.8615855,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.720576286315918
    },
    {
      "auxiliary_loss_clip": 0.01191628,
      "auxiliary_loss_mlp": 0.00903731,
      "balance_loss_clip": 1.06014276,
      "balance_loss_mlp": 1.00252676,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 2.1079850923208796,
      "language_loss": 0.74514604,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76609969,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.783129930496216
    },
    {
      "auxiliary_loss_clip": 0.01174198,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.05267692,
      "balance_loss_mlp": 1.02207506,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 4.107150300857483,
      "language_loss": 0.81807804,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84013414,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.751116991043091
    },
    {
      "auxiliary_loss_clip": 0.01182633,
      "auxiliary_loss_mlp": 0.01031537,
      "balance_loss_clip": 1.05835986,
      "balance_loss_mlp": 1.02201843,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.9970603542819045,
      "language_loss": 0.8106342,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.83277595,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.8142507076263428
    },
    {
      "auxiliary_loss_clip": 0.01182771,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.05593288,
      "balance_loss_mlp": 1.02115369,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 3.5484590966202396,
      "language_loss": 0.7628333,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78496444,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.693052291870117
    },
    {
      "auxiliary_loss_clip": 0.01165708,
      "auxiliary_loss_mlp": 0.01033091,
      "balance_loss_clip": 1.05573773,
      "balance_loss_mlp": 1.02410889,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 3.4629344822348047,
      "language_loss": 0.78569984,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80768788,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.7317087650299072
    },
    {
      "auxiliary_loss_clip": 0.01173729,
      "auxiliary_loss_mlp": 0.01030702,
      "balance_loss_clip": 1.05153537,
      "balance_loss_mlp": 1.02193451,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.2516170695140754,
      "language_loss": 0.68735176,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.70939606,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.7773218154907227
    },
    {
      "auxiliary_loss_clip": 0.01200226,
      "auxiliary_loss_mlp": 0.0090247,
      "balance_loss_clip": 1.05829167,
      "balance_loss_mlp": 1.00258887,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 2.3040205618735756,
      "language_loss": 0.70835632,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.72938335,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.634385347366333
    },
    {
      "auxiliary_loss_clip": 0.01201421,
      "auxiliary_loss_mlp": 0.01031674,
      "balance_loss_clip": 1.05767119,
      "balance_loss_mlp": 1.02144551,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 3.741201965998887,
      "language_loss": 0.7733115,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79564244,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.6574044227600098
    },
    {
      "auxiliary_loss_clip": 0.01205784,
      "auxiliary_loss_mlp": 0.01029519,
      "balance_loss_clip": 1.06398988,
      "balance_loss_mlp": 1.01986909,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.7209577285051436,
      "language_loss": 0.72897887,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.75133193,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.685194492340088
    },
    {
      "auxiliary_loss_clip": 0.01206296,
      "auxiliary_loss_mlp": 0.01032708,
      "balance_loss_clip": 1.05981946,
      "balance_loss_mlp": 1.02345121,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 2.1327680344185107,
      "language_loss": 0.76565111,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78804111,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.8621439933776855
    },
    {
      "auxiliary_loss_clip": 0.01212177,
      "auxiliary_loss_mlp": 0.01035883,
      "balance_loss_clip": 1.06180859,
      "balance_loss_mlp": 1.02601254,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 2.4588936266111427,
      "language_loss": 0.783301,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80578148,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 3.5673511028289795
    },
    {
      "auxiliary_loss_clip": 0.01184545,
      "auxiliary_loss_mlp": 0.01033315,
      "balance_loss_clip": 1.05729771,
      "balance_loss_mlp": 1.02332509,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.492305374850275,
      "language_loss": 0.83710158,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85928011,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 3.595918893814087
    },
    {
      "auxiliary_loss_clip": 0.01206656,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.05786157,
      "balance_loss_mlp": 1.021052,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.3442952195013547,
      "language_loss": 0.8187654,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84114742,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.646883964538574
    },
    {
      "auxiliary_loss_clip": 0.01208943,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.06229734,
      "balance_loss_mlp": 1.02524006,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 2.1492840196842877,
      "language_loss": 0.82717782,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84960824,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 3.5786421298980713
    },
    {
      "auxiliary_loss_clip": 0.01163723,
      "auxiliary_loss_mlp": 0.01027219,
      "balance_loss_clip": 1.05241108,
      "balance_loss_mlp": 1.01801634,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 2.906467144725269,
      "language_loss": 0.7426393,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76454872,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.7388458251953125
    },
    {
      "auxiliary_loss_clip": 0.01189013,
      "auxiliary_loss_mlp": 0.0103198,
      "balance_loss_clip": 1.05497968,
      "balance_loss_mlp": 1.02240181,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.3279357194593553,
      "language_loss": 0.77995622,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.8021661,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.8748996257781982
    },
    {
      "auxiliary_loss_clip": 0.01191281,
      "auxiliary_loss_mlp": 0.00904177,
      "balance_loss_clip": 1.05997086,
      "balance_loss_mlp": 1.00271058,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.3668735178709044,
      "language_loss": 0.80563354,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82658815,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.6929147243499756
    },
    {
      "auxiliary_loss_clip": 0.01205533,
      "auxiliary_loss_mlp": 0.01033154,
      "balance_loss_clip": 1.05979693,
      "balance_loss_mlp": 1.02340817,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 11.802996302313,
      "language_loss": 0.8316257,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85401261,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 3.513044834136963
    },
    {
      "auxiliary_loss_clip": 0.01198089,
      "auxiliary_loss_mlp": 0.010328,
      "balance_loss_clip": 1.05700672,
      "balance_loss_mlp": 1.02372205,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 5.783976621586013,
      "language_loss": 0.7316525,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75396144,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.675786256790161
    },
    {
      "auxiliary_loss_clip": 0.01188671,
      "auxiliary_loss_mlp": 0.01032052,
      "balance_loss_clip": 1.05945897,
      "balance_loss_mlp": 1.02289665,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.1743026486465666,
      "language_loss": 0.69867182,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72087902,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.6962907314300537
    },
    {
      "auxiliary_loss_clip": 0.01208111,
      "auxiliary_loss_mlp": 0.01029795,
      "balance_loss_clip": 1.05953276,
      "balance_loss_mlp": 1.0206759,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 2.2612726318734313,
      "language_loss": 0.83757704,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.85995615,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.611178159713745
    },
    {
      "auxiliary_loss_clip": 0.01192006,
      "auxiliary_loss_mlp": 0.01026659,
      "balance_loss_clip": 1.05748558,
      "balance_loss_mlp": 1.0168128,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.9702014689647789,
      "language_loss": 0.83142561,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85361224,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.678293466567993
    },
    {
      "auxiliary_loss_clip": 0.01214244,
      "auxiliary_loss_mlp": 0.01037484,
      "balance_loss_clip": 1.06284595,
      "balance_loss_mlp": 1.02803099,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 3.768258738239983,
      "language_loss": 0.76634574,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78886306,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.5768020153045654
    },
    {
      "auxiliary_loss_clip": 0.01153047,
      "auxiliary_loss_mlp": 0.01031421,
      "balance_loss_clip": 1.05928063,
      "balance_loss_mlp": 1.02908444,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9329231038642019,
      "language_loss": 0.60570079,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62754548,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 2.84393048286438
    },
    {
      "auxiliary_loss_clip": 0.0116774,
      "auxiliary_loss_mlp": 0.01036527,
      "balance_loss_clip": 1.05289614,
      "balance_loss_mlp": 1.02669263,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 1.9052418857660984,
      "language_loss": 0.75041318,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77245581,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.6119577884674072
    },
    {
      "auxiliary_loss_clip": 0.01186915,
      "auxiliary_loss_mlp": 0.01034395,
      "balance_loss_clip": 1.05858469,
      "balance_loss_mlp": 1.0240953,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 2.1976757978238584,
      "language_loss": 0.87751716,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.89973027,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.689204692840576
    },
    {
      "auxiliary_loss_clip": 0.0117518,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.0564959,
      "balance_loss_mlp": 1.0262785,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 3.0656408025526543,
      "language_loss": 0.7745474,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79665929,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.6276559829711914
    },
    {
      "auxiliary_loss_clip": 0.01207751,
      "auxiliary_loss_mlp": 0.01033437,
      "balance_loss_clip": 1.06474018,
      "balance_loss_mlp": 1.02441335,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.8261074210965913,
      "language_loss": 0.86965585,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89206779,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.715970993041992
    },
    {
      "auxiliary_loss_clip": 0.01191081,
      "auxiliary_loss_mlp": 0.01029113,
      "balance_loss_clip": 1.05880129,
      "balance_loss_mlp": 1.01966572,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.083263138739635,
      "language_loss": 0.88123894,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90344083,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.700059175491333
    },
    {
      "auxiliary_loss_clip": 0.01180427,
      "auxiliary_loss_mlp": 0.01035611,
      "balance_loss_clip": 1.05228925,
      "balance_loss_mlp": 1.02616322,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 3.253679590151401,
      "language_loss": 0.61482871,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63698912,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.6788156032562256
    },
    {
      "auxiliary_loss_clip": 0.01196413,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.05937028,
      "balance_loss_mlp": 1.02203655,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 3.222359786005131,
      "language_loss": 0.8236829,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84597051,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.6084444522857666
    },
    {
      "auxiliary_loss_clip": 0.01186251,
      "auxiliary_loss_mlp": 0.01036978,
      "balance_loss_clip": 1.05710101,
      "balance_loss_mlp": 1.02708983,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 1.945215977798969,
      "language_loss": 0.73444915,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75668144,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.675198554992676
    },
    {
      "auxiliary_loss_clip": 0.01183822,
      "auxiliary_loss_mlp": 0.01033236,
      "balance_loss_clip": 1.05726147,
      "balance_loss_mlp": 1.0229063,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 4.22901515515871,
      "language_loss": 0.85602438,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87819493,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.619605541229248
    },
    {
      "auxiliary_loss_clip": 0.01188338,
      "auxiliary_loss_mlp": 0.01045444,
      "balance_loss_clip": 1.06827593,
      "balance_loss_mlp": 1.0362947,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.924080072876901,
      "language_loss": 0.81797922,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84031701,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.6602089405059814
    },
    {
      "auxiliary_loss_clip": 0.01216913,
      "auxiliary_loss_mlp": 0.01032059,
      "balance_loss_clip": 1.06356359,
      "balance_loss_mlp": 1.02159214,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 3.095723492518925,
      "language_loss": 0.77681005,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79929978,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.6028358936309814
    },
    {
      "auxiliary_loss_clip": 0.01190727,
      "auxiliary_loss_mlp": 0.01033239,
      "balance_loss_clip": 1.05759978,
      "balance_loss_mlp": 1.02353537,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 5.79628675601733,
      "language_loss": 0.80182081,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82406044,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.7493863105773926
    },
    {
      "auxiliary_loss_clip": 0.01183046,
      "auxiliary_loss_mlp": 0.01031849,
      "balance_loss_clip": 1.05902493,
      "balance_loss_mlp": 1.02291465,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 2.0457984660021897,
      "language_loss": 0.74049872,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76264763,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 3.662259578704834
    },
    {
      "auxiliary_loss_clip": 0.01212822,
      "auxiliary_loss_mlp": 0.01036934,
      "balance_loss_clip": 1.06307697,
      "balance_loss_mlp": 1.0274806,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 6.681267342923919,
      "language_loss": 0.76014125,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78263879,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.574496030807495
    },
    {
      "auxiliary_loss_clip": 0.01193541,
      "auxiliary_loss_mlp": 0.01032945,
      "balance_loss_clip": 1.06172192,
      "balance_loss_mlp": 1.02318799,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.204155526170683,
      "language_loss": 0.81052339,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.83278829,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 3.6396543979644775
    },
    {
      "auxiliary_loss_clip": 0.01188575,
      "auxiliary_loss_mlp": 0.01034677,
      "balance_loss_clip": 1.056831,
      "balance_loss_mlp": 1.024544,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 3.8277422888474333,
      "language_loss": 0.71234083,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73457336,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 3.6742606163024902
    },
    {
      "auxiliary_loss_clip": 0.01197486,
      "auxiliary_loss_mlp": 0.01034561,
      "balance_loss_clip": 1.06064308,
      "balance_loss_mlp": 1.02519119,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 2.7849014658602083,
      "language_loss": 0.79206216,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81438261,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.6080682277679443
    },
    {
      "auxiliary_loss_clip": 0.01186841,
      "auxiliary_loss_mlp": 0.01040747,
      "balance_loss_clip": 1.06160426,
      "balance_loss_mlp": 1.03251588,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 1.805177439442,
      "language_loss": 0.88253605,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90481186,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.8018815517425537
    },
    {
      "auxiliary_loss_clip": 0.01176693,
      "auxiliary_loss_mlp": 0.01034999,
      "balance_loss_clip": 1.05775177,
      "balance_loss_mlp": 1.02590966,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 2.3449693125909294,
      "language_loss": 0.82916367,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85128057,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.7066657543182373
    },
    {
      "auxiliary_loss_clip": 0.01107323,
      "auxiliary_loss_mlp": 0.00895454,
      "balance_loss_clip": 1.03845084,
      "balance_loss_mlp": 1.0020144,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8812881579511509,
      "language_loss": 0.62735331,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64738107,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.2171459197998047
    },
    {
      "auxiliary_loss_clip": 0.01199518,
      "auxiliary_loss_mlp": 0.00903513,
      "balance_loss_clip": 1.06398082,
      "balance_loss_mlp": 1.002918,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.52721059153826,
      "language_loss": 0.77424389,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.7952742,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 3.6221566200256348
    },
    {
      "auxiliary_loss_clip": 0.01206609,
      "auxiliary_loss_mlp": 0.01029242,
      "balance_loss_clip": 1.06237173,
      "balance_loss_mlp": 1.02005696,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 7.141200866061443,
      "language_loss": 0.88565779,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.90801632,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.6725828647613525
    },
    {
      "auxiliary_loss_clip": 0.01218561,
      "auxiliary_loss_mlp": 0.01034382,
      "balance_loss_clip": 1.06538916,
      "balance_loss_mlp": 1.02354634,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.9771490309558946,
      "language_loss": 0.69437897,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71690845,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.778855323791504
    },
    {
      "auxiliary_loss_clip": 0.011298,
      "auxiliary_loss_mlp": 0.01021879,
      "balance_loss_clip": 1.04783452,
      "balance_loss_mlp": 1.01956606,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7695782685140683,
      "language_loss": 0.5441215,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.5656383,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.061539888381958
    },
    {
      "auxiliary_loss_clip": 0.01188831,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.05844188,
      "balance_loss_mlp": 1.01844513,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.120915017962802,
      "language_loss": 0.80351281,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.82568997,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.6896424293518066
    },
    {
      "auxiliary_loss_clip": 0.01191713,
      "auxiliary_loss_mlp": 0.01034709,
      "balance_loss_clip": 1.05799651,
      "balance_loss_mlp": 1.02488589,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 2.9992125626128225,
      "language_loss": 0.80718136,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.8294456,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.6981875896453857
    },
    {
      "auxiliary_loss_clip": 0.01215986,
      "auxiliary_loss_mlp": 0.01039721,
      "balance_loss_clip": 1.06515443,
      "balance_loss_mlp": 1.02983832,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.341104977518727,
      "language_loss": 0.76401699,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78657401,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.584484577178955
    },
    {
      "auxiliary_loss_clip": 0.01174102,
      "auxiliary_loss_mlp": 0.01033751,
      "balance_loss_clip": 1.06005168,
      "balance_loss_mlp": 1.0238328,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.7192985433548413,
      "language_loss": 0.66510117,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68717968,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.7183797359466553
    },
    {
      "auxiliary_loss_clip": 0.0120862,
      "auxiliary_loss_mlp": 0.01038478,
      "balance_loss_clip": 1.060812,
      "balance_loss_mlp": 1.02866137,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 2.3729036501146457,
      "language_loss": 0.7154237,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.73789465,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.63364839553833
    },
    {
      "auxiliary_loss_clip": 0.01180739,
      "auxiliary_loss_mlp": 0.01032368,
      "balance_loss_clip": 1.0584271,
      "balance_loss_mlp": 1.02318239,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 3.3642415732865496,
      "language_loss": 0.78637701,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80850804,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.692319869995117
    },
    {
      "auxiliary_loss_clip": 0.01139849,
      "auxiliary_loss_mlp": 0.01032565,
      "balance_loss_clip": 1.04634285,
      "balance_loss_mlp": 1.02328432,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 3.5597262032856314,
      "language_loss": 0.79717207,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81889617,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.8493096828460693
    },
    {
      "auxiliary_loss_clip": 0.01207482,
      "auxiliary_loss_mlp": 0.01037496,
      "balance_loss_clip": 1.06228042,
      "balance_loss_mlp": 1.02775049,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 2.9047905577276816,
      "language_loss": 0.89866173,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92111146,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.6208512783050537
    },
    {
      "auxiliary_loss_clip": 0.01160775,
      "auxiliary_loss_mlp": 0.01036228,
      "balance_loss_clip": 1.05402088,
      "balance_loss_mlp": 1.02719831,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 3.8472263430009583,
      "language_loss": 0.81279629,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83476633,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.720536947250366
    },
    {
      "auxiliary_loss_clip": 0.01211729,
      "auxiliary_loss_mlp": 0.01034199,
      "balance_loss_clip": 1.0632906,
      "balance_loss_mlp": 1.024966,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 2.7089272021923647,
      "language_loss": 0.66902351,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.69148278,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.6811914443969727
    },
    {
      "auxiliary_loss_clip": 0.01170268,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.05116153,
      "balance_loss_mlp": 1.01984739,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.9220889496053164,
      "language_loss": 0.7874015,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.80939698,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.7154252529144287
    },
    {
      "auxiliary_loss_clip": 0.01191122,
      "auxiliary_loss_mlp": 0.01031232,
      "balance_loss_clip": 1.06381595,
      "balance_loss_mlp": 1.02213669,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.8304368314029684,
      "language_loss": 0.67185318,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.69407672,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.7035958766937256
    },
    {
      "auxiliary_loss_clip": 0.01203205,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.06387472,
      "balance_loss_mlp": 1.02109921,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 2.068315356311098,
      "language_loss": 0.77023077,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79256535,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.591993808746338
    },
    {
      "auxiliary_loss_clip": 0.01112203,
      "auxiliary_loss_mlp": 0.01015284,
      "balance_loss_clip": 1.05510235,
      "balance_loss_mlp": 1.0132103,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.9114143755680872,
      "language_loss": 0.62651765,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64779246,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.392580270767212
    },
    {
      "auxiliary_loss_clip": 0.01191623,
      "auxiliary_loss_mlp": 0.01032278,
      "balance_loss_clip": 1.05875909,
      "balance_loss_mlp": 1.02276468,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.4381725462949904,
      "language_loss": 1.00783956,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.03007853,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.7237558364868164
    },
    {
      "auxiliary_loss_clip": 0.01198903,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.06005681,
      "balance_loss_mlp": 1.01955819,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 3.3737704243232023,
      "language_loss": 0.78820586,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.8104828,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 3.603170156478882
    },
    {
      "auxiliary_loss_clip": 0.0119053,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.06401753,
      "balance_loss_mlp": 1.02097309,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.531574706179668,
      "language_loss": 0.87798524,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.900195,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.663137435913086
    },
    {
      "auxiliary_loss_clip": 0.01197402,
      "auxiliary_loss_mlp": 0.01039313,
      "balance_loss_clip": 1.05967212,
      "balance_loss_mlp": 1.02960372,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.1823652887036205,
      "language_loss": 0.76099604,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78336322,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 4.554541826248169
    },
    {
      "auxiliary_loss_clip": 0.01179013,
      "auxiliary_loss_mlp": 0.01035329,
      "balance_loss_clip": 1.05643749,
      "balance_loss_mlp": 1.02600706,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.2329863046006877,
      "language_loss": 0.90824127,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93038476,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.6332883834838867
    },
    {
      "auxiliary_loss_clip": 0.01213898,
      "auxiliary_loss_mlp": 0.01034619,
      "balance_loss_clip": 1.06518936,
      "balance_loss_mlp": 1.02521944,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 2.879654457598008,
      "language_loss": 0.88559175,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90807694,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.659212350845337
    },
    {
      "auxiliary_loss_clip": 0.01189181,
      "auxiliary_loss_mlp": 0.01029883,
      "balance_loss_clip": 1.05997467,
      "balance_loss_mlp": 1.02066779,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.674509409474904,
      "language_loss": 0.77882981,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80102044,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.7428884506225586
    },
    {
      "auxiliary_loss_clip": 0.0110817,
      "auxiliary_loss_mlp": 0.01004571,
      "balance_loss_clip": 1.0418303,
      "balance_loss_mlp": 1.00263965,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6670822733875442,
      "language_loss": 0.56805158,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.58917904,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.327317237854004
    },
    {
      "auxiliary_loss_clip": 0.01173114,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.05538321,
      "balance_loss_mlp": 1.0236752,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.925865154904201,
      "language_loss": 0.83454925,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85660386,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.7580597400665283
    },
    {
      "auxiliary_loss_clip": 0.01191338,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.0603019,
      "balance_loss_mlp": 1.01655698,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 2.4990565369952398,
      "language_loss": 0.81303722,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83520889,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 3.6142733097076416
    },
    {
      "auxiliary_loss_clip": 0.01189408,
      "auxiliary_loss_mlp": 0.01028651,
      "balance_loss_clip": 1.0589664,
      "balance_loss_mlp": 1.0198946,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.1917834953844983,
      "language_loss": 0.8582117,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88039225,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.6190760135650635
    },
    {
      "auxiliary_loss_clip": 0.01158097,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.05253577,
      "balance_loss_mlp": 1.01879299,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 2.3515818807787943,
      "language_loss": 0.71220577,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73406661,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.1101267337799072
    },
    {
      "auxiliary_loss_clip": 0.01173479,
      "auxiliary_loss_mlp": 0.01028673,
      "balance_loss_clip": 1.05675435,
      "balance_loss_mlp": 1.02003074,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.8284722972032523,
      "language_loss": 0.82123291,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84325445,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.776416063308716
    },
    {
      "auxiliary_loss_clip": 0.01090717,
      "auxiliary_loss_mlp": 0.01005327,
      "balance_loss_clip": 1.03170466,
      "balance_loss_mlp": 1.00326502,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9193133910544666,
      "language_loss": 0.56838012,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58934057,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.362605333328247
    },
    {
      "auxiliary_loss_clip": 0.01205044,
      "auxiliary_loss_mlp": 0.01036635,
      "balance_loss_clip": 1.06174016,
      "balance_loss_mlp": 1.02618599,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.4465007371716254,
      "language_loss": 0.7127223,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.73513913,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.5637545585632324
    },
    {
      "auxiliary_loss_clip": 0.01181744,
      "auxiliary_loss_mlp": 0.01039084,
      "balance_loss_clip": 1.05532479,
      "balance_loss_mlp": 1.02902913,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.628266967368191,
      "language_loss": 0.67489785,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69710606,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.6720964908599854
    },
    {
      "auxiliary_loss_clip": 0.01182344,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.05435967,
      "balance_loss_mlp": 1.02213788,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 2.0564754114429173,
      "language_loss": 0.79845834,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82060564,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.701246738433838
    },
    {
      "auxiliary_loss_clip": 0.01199474,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.05925417,
      "balance_loss_mlp": 1.02295995,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.8459113350456944,
      "language_loss": 0.75726557,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.77958268,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.6696434020996094
    },
    {
      "auxiliary_loss_clip": 0.01188719,
      "auxiliary_loss_mlp": 0.00903519,
      "balance_loss_clip": 1.05752206,
      "balance_loss_mlp": 1.00257564,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.2517763788195766,
      "language_loss": 0.82474375,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84566617,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.675476551055908
    },
    {
      "auxiliary_loss_clip": 0.01192407,
      "auxiliary_loss_mlp": 0.01030771,
      "balance_loss_clip": 1.05741954,
      "balance_loss_mlp": 1.02133501,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.283553083557221,
      "language_loss": 0.89947122,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.92170298,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.660670518875122
    },
    {
      "auxiliary_loss_clip": 0.01189984,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.06048036,
      "balance_loss_mlp": 1.02005911,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 2.1946743855572426,
      "language_loss": 0.75742412,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.77962673,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.662276029586792
    },
    {
      "auxiliary_loss_clip": 0.01202965,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.05959964,
      "balance_loss_mlp": 1.02076483,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 3.9826808298160237,
      "language_loss": 0.84775001,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87007803,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.6448400020599365
    },
    {
      "auxiliary_loss_clip": 0.01203784,
      "auxiliary_loss_mlp": 0.01036054,
      "balance_loss_clip": 1.05889821,
      "balance_loss_mlp": 1.02617192,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 2.079585246441309,
      "language_loss": 0.84386039,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86625874,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.6612887382507324
    },
    {
      "auxiliary_loss_clip": 0.01174502,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.05112743,
      "balance_loss_mlp": 1.0266149,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 2.437047532296091,
      "language_loss": 0.78969741,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.8118062,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.7407774925231934
    },
    {
      "auxiliary_loss_clip": 0.01207435,
      "auxiliary_loss_mlp": 0.01037458,
      "balance_loss_clip": 1.0593816,
      "balance_loss_mlp": 1.02753401,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 4.189142025162961,
      "language_loss": 0.80938041,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83182943,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.6705381870269775
    },
    {
      "auxiliary_loss_clip": 0.01155271,
      "auxiliary_loss_mlp": 0.01035106,
      "balance_loss_clip": 1.05150139,
      "balance_loss_mlp": 1.02617145,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 2.445184607336515,
      "language_loss": 0.76443422,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.78633797,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.78957200050354
    },
    {
      "auxiliary_loss_clip": 0.01211482,
      "auxiliary_loss_mlp": 0.01034925,
      "balance_loss_clip": 1.06157005,
      "balance_loss_mlp": 1.02466667,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 3.7737397689023613,
      "language_loss": 0.83773887,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.86020291,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.6584463119506836
    },
    {
      "auxiliary_loss_clip": 0.01188681,
      "auxiliary_loss_mlp": 0.01032719,
      "balance_loss_clip": 1.05756843,
      "balance_loss_mlp": 1.02325976,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 3.474714403663279,
      "language_loss": 0.80073547,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82294953,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.7085254192352295
    },
    {
      "auxiliary_loss_clip": 0.01187803,
      "auxiliary_loss_mlp": 0.01030689,
      "balance_loss_clip": 1.05726564,
      "balance_loss_mlp": 1.02161741,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 2.2597101650355174,
      "language_loss": 0.9029963,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92518127,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 3.796152353286743
    },
    {
      "auxiliary_loss_clip": 0.01180601,
      "auxiliary_loss_mlp": 0.01033997,
      "balance_loss_clip": 1.05560887,
      "balance_loss_mlp": 1.02444839,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 2.1411111811649732,
      "language_loss": 0.79258811,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.8147341,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 3.614325523376465
    },
    {
      "auxiliary_loss_clip": 0.01184761,
      "auxiliary_loss_mlp": 0.01032473,
      "balance_loss_clip": 1.05742693,
      "balance_loss_mlp": 1.0228411,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 2.0821182609128344,
      "language_loss": 0.80865657,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.83082891,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 3.5501420497894287
    },
    {
      "auxiliary_loss_clip": 0.01105589,
      "auxiliary_loss_mlp": 0.01007613,
      "balance_loss_clip": 1.030864,
      "balance_loss_mlp": 1.00572991,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9712257565416811,
      "language_loss": 0.58170557,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60283756,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.2100772857666016
    },
    {
      "auxiliary_loss_clip": 0.01201277,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.06062543,
      "balance_loss_mlp": 1.01994705,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 2.090472609995545,
      "language_loss": 0.70692509,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.72923231,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.64827036857605
    },
    {
      "auxiliary_loss_clip": 0.01184489,
      "auxiliary_loss_mlp": 0.01030023,
      "balance_loss_clip": 1.05518878,
      "balance_loss_mlp": 1.02009249,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.8064414359059469,
      "language_loss": 0.80034214,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82248724,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.713035821914673
    },
    {
      "auxiliary_loss_clip": 0.01176794,
      "auxiliary_loss_mlp": 0.01037101,
      "balance_loss_clip": 1.05518627,
      "balance_loss_mlp": 1.02765393,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.1812745706170698,
      "language_loss": 0.70492363,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72706258,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.7802529335021973
    },
    {
      "auxiliary_loss_clip": 0.01119568,
      "auxiliary_loss_mlp": 0.01005803,
      "balance_loss_clip": 1.03733301,
      "balance_loss_mlp": 1.00365686,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.8996277636846192,
      "language_loss": 0.63437897,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65563267,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 4.089505195617676
    },
    {
      "auxiliary_loss_clip": 0.01123708,
      "auxiliary_loss_mlp": 0.01005371,
      "balance_loss_clip": 1.03751814,
      "balance_loss_mlp": 1.00326061,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8214343954617211,
      "language_loss": 0.55788672,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.5791775,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.2436139583587646
    },
    {
      "auxiliary_loss_clip": 0.01203482,
      "auxiliary_loss_mlp": 0.01034278,
      "balance_loss_clip": 1.06180036,
      "balance_loss_mlp": 1.02498579,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.7024013705910397,
      "language_loss": 0.74600667,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.76838422,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.697014093399048
    },
    {
      "auxiliary_loss_clip": 0.0119003,
      "auxiliary_loss_mlp": 0.01024769,
      "balance_loss_clip": 1.05705142,
      "balance_loss_mlp": 1.01613879,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 3.6870615014018013,
      "language_loss": 0.83877254,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86092055,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.6792562007904053
    },
    {
      "auxiliary_loss_clip": 0.01209342,
      "auxiliary_loss_mlp": 0.010356,
      "balance_loss_clip": 1.05943131,
      "balance_loss_mlp": 1.0255152,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 2.062427403570692,
      "language_loss": 0.7954033,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81785274,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.6320950984954834
    },
    {
      "auxiliary_loss_clip": 0.01164911,
      "auxiliary_loss_mlp": 0.01031539,
      "balance_loss_clip": 1.05464399,
      "balance_loss_mlp": 1.02244973,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.7988901786773726,
      "language_loss": 0.70909286,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73105735,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.727221727371216
    },
    {
      "auxiliary_loss_clip": 0.01208433,
      "auxiliary_loss_mlp": 0.01032212,
      "balance_loss_clip": 1.06094086,
      "balance_loss_mlp": 1.02266884,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 4.933815579960102,
      "language_loss": 0.80017459,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.82258099,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.621697187423706
    },
    {
      "auxiliary_loss_clip": 0.01189787,
      "auxiliary_loss_mlp": 0.01040379,
      "balance_loss_clip": 1.05540216,
      "balance_loss_mlp": 1.03093135,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 2.1974657421421275,
      "language_loss": 0.81973219,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.8420338,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.669163942337036
    },
    {
      "auxiliary_loss_clip": 0.0120426,
      "auxiliary_loss_mlp": 0.01031966,
      "balance_loss_clip": 1.05597782,
      "balance_loss_mlp": 1.02198851,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 7.095937823109682,
      "language_loss": 0.76631033,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78867263,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.637355327606201
    },
    {
      "auxiliary_loss_clip": 0.011829,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.05648375,
      "balance_loss_mlp": 1.02011204,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.638769580474827,
      "language_loss": 0.6857121,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70784724,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.70926570892334
    },
    {
      "auxiliary_loss_clip": 0.01198092,
      "auxiliary_loss_mlp": 0.0103521,
      "balance_loss_clip": 1.06218946,
      "balance_loss_mlp": 1.02637017,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.649234524925165,
      "language_loss": 0.82087791,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84321094,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.676184892654419
    },
    {
      "auxiliary_loss_clip": 0.01186523,
      "auxiliary_loss_mlp": 0.0090377,
      "balance_loss_clip": 1.05589747,
      "balance_loss_mlp": 1.0024848,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.6267906787423074,
      "language_loss": 0.78831881,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80922174,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.7076547145843506
    },
    {
      "auxiliary_loss_clip": 0.01187695,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.05627155,
      "balance_loss_mlp": 1.02112854,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 2.094964959494725,
      "language_loss": 0.80148566,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82366145,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.6726224422454834
    },
    {
      "auxiliary_loss_clip": 0.01083275,
      "auxiliary_loss_mlp": 0.0101143,
      "balance_loss_clip": 1.02688134,
      "balance_loss_mlp": 1.00932026,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 1.2426445176305627,
      "language_loss": 0.57836461,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59931159,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.379746913909912
    },
    {
      "auxiliary_loss_clip": 0.01192411,
      "auxiliary_loss_mlp": 0.01031563,
      "balance_loss_clip": 1.05707192,
      "balance_loss_mlp": 1.02250314,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.7649503991072941,
      "language_loss": 0.86502838,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88726813,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.747199773788452
    },
    {
      "auxiliary_loss_clip": 0.01198195,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.05641603,
      "balance_loss_mlp": 1.02049232,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 2.148865682497165,
      "language_loss": 0.84829938,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87057436,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.65502667427063
    },
    {
      "auxiliary_loss_clip": 0.01171353,
      "auxiliary_loss_mlp": 0.01036153,
      "balance_loss_clip": 1.05070293,
      "balance_loss_mlp": 1.02650952,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 3.000679451642658,
      "language_loss": 0.78845799,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.81053305,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.6412301063537598
    },
    {
      "auxiliary_loss_clip": 0.01203821,
      "auxiliary_loss_mlp": 0.01033788,
      "balance_loss_clip": 1.05664754,
      "balance_loss_mlp": 1.02390003,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 2.1316747279943855,
      "language_loss": 0.80274534,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.8251214,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.677029848098755
    },
    {
      "auxiliary_loss_clip": 0.01210648,
      "auxiliary_loss_mlp": 0.01039133,
      "balance_loss_clip": 1.06279516,
      "balance_loss_mlp": 1.02925038,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 2.68474910260398,
      "language_loss": 0.78460968,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80710745,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.6535353660583496
    },
    {
      "auxiliary_loss_clip": 0.01183281,
      "auxiliary_loss_mlp": 0.01031712,
      "balance_loss_clip": 1.05641985,
      "balance_loss_mlp": 1.02282524,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.6733693881160205,
      "language_loss": 0.6883381,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71048808,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.800855875015259
    },
    {
      "auxiliary_loss_clip": 0.01159318,
      "auxiliary_loss_mlp": 0.01030754,
      "balance_loss_clip": 1.04908228,
      "balance_loss_mlp": 1.02078855,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.8390065625165166,
      "language_loss": 0.77461183,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.7965126,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 4.014597177505493
    },
    {
      "auxiliary_loss_clip": 0.01195722,
      "auxiliary_loss_mlp": 0.01032239,
      "balance_loss_clip": 1.05970728,
      "balance_loss_mlp": 1.02384627,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.8096856663161238,
      "language_loss": 0.84913188,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87141144,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 3.5863962173461914
    },
    {
      "auxiliary_loss_clip": 0.01112796,
      "auxiliary_loss_mlp": 0.01001117,
      "balance_loss_clip": 1.03633273,
      "balance_loss_mlp": 0.99889994,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7838932999115262,
      "language_loss": 0.63171935,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.6528585,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.1667277812957764
    },
    {
      "auxiliary_loss_clip": 0.01176695,
      "auxiliary_loss_mlp": 0.01032985,
      "balance_loss_clip": 1.05396748,
      "balance_loss_mlp": 1.02317405,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 2.0434577584964595,
      "language_loss": 0.7769264,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.79902321,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.7413017749786377
    },
    {
      "auxiliary_loss_clip": 0.01183582,
      "auxiliary_loss_mlp": 0.01031671,
      "balance_loss_clip": 1.05431521,
      "balance_loss_mlp": 1.02183032,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.261161923709583,
      "language_loss": 0.83877099,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86092353,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.7100391387939453
    },
    {
      "auxiliary_loss_clip": 0.01190007,
      "auxiliary_loss_mlp": 0.01034823,
      "balance_loss_clip": 1.05279446,
      "balance_loss_mlp": 1.02551889,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 4.818275641404954,
      "language_loss": 0.80826533,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.83051354,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.723695755004883
    },
    {
      "auxiliary_loss_clip": 0.01182487,
      "auxiliary_loss_mlp": 0.01029953,
      "balance_loss_clip": 1.05312848,
      "balance_loss_mlp": 1.02030873,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 2.0173690403113236,
      "language_loss": 0.86780912,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88993347,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.7608721256256104
    },
    {
      "auxiliary_loss_clip": 0.01178573,
      "auxiliary_loss_mlp": 0.01032032,
      "balance_loss_clip": 1.05534959,
      "balance_loss_mlp": 1.02275145,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.7372205514358308,
      "language_loss": 0.91308618,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93519223,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 3.692974090576172
    },
    {
      "auxiliary_loss_clip": 0.01182172,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.05610085,
      "balance_loss_mlp": 1.02197993,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.098057826286818,
      "language_loss": 0.78146493,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80359286,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.6779398918151855
    },
    {
      "auxiliary_loss_clip": 0.01207708,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.06135738,
      "balance_loss_mlp": 1.02347445,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 2.04327344942164,
      "language_loss": 0.78281558,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80521441,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.5679500102996826
    },
    {
      "auxiliary_loss_clip": 0.01188112,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.05653739,
      "balance_loss_mlp": 1.02053547,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 2.3199164445623226,
      "language_loss": 0.72591913,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.74809611,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.661888837814331
    },
    {
      "auxiliary_loss_clip": 0.01172793,
      "auxiliary_loss_mlp": 0.0103467,
      "balance_loss_clip": 1.05162776,
      "balance_loss_mlp": 1.02555084,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.8592464488564187,
      "language_loss": 0.60177791,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62385255,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.6335153579711914
    },
    {
      "auxiliary_loss_clip": 0.0118676,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.05611634,
      "balance_loss_mlp": 1.02105951,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.281792657304648,
      "language_loss": 0.75016332,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77233279,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.6772706508636475
    },
    {
      "auxiliary_loss_clip": 0.01181492,
      "auxiliary_loss_mlp": 0.01030244,
      "balance_loss_clip": 1.05530596,
      "balance_loss_mlp": 1.02121997,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.6789379732762777,
      "language_loss": 0.87410748,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89622486,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.7293853759765625
    },
    {
      "auxiliary_loss_clip": 0.01153306,
      "auxiliary_loss_mlp": 0.00903211,
      "balance_loss_clip": 1.05093467,
      "balance_loss_mlp": 1.00220287,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 1.9598507163046095,
      "language_loss": 0.75025982,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77082503,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.8644888401031494
    },
    {
      "auxiliary_loss_clip": 0.01158075,
      "auxiliary_loss_mlp": 0.00903326,
      "balance_loss_clip": 1.04806757,
      "balance_loss_mlp": 1.00224662,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 1.9679005148394142,
      "language_loss": 0.74054861,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76116264,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.795523166656494
    },
    {
      "auxiliary_loss_clip": 0.01190454,
      "auxiliary_loss_mlp": 0.01033233,
      "balance_loss_clip": 1.05527163,
      "balance_loss_mlp": 1.02370262,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.4372321591014687,
      "language_loss": 0.83576846,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.8580054,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.627589702606201
    },
    {
      "auxiliary_loss_clip": 0.01200452,
      "auxiliary_loss_mlp": 0.00903181,
      "balance_loss_clip": 1.05622149,
      "balance_loss_mlp": 1.00239789,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 4.104955956442589,
      "language_loss": 0.80291891,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82395518,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.7337512969970703
    },
    {
      "auxiliary_loss_clip": 0.01183719,
      "auxiliary_loss_mlp": 0.01032686,
      "balance_loss_clip": 1.05382061,
      "balance_loss_mlp": 1.02351284,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 2.16973951581946,
      "language_loss": 0.84179211,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86395609,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.634580612182617
    },
    {
      "auxiliary_loss_clip": 0.01124056,
      "auxiliary_loss_mlp": 0.01004988,
      "balance_loss_clip": 1.03720939,
      "balance_loss_mlp": 1.00284243,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.78890287492873,
      "language_loss": 0.52722692,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54851735,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.219109535217285
    },
    {
      "auxiliary_loss_clip": 0.01171892,
      "auxiliary_loss_mlp": 0.010346,
      "balance_loss_clip": 1.05323434,
      "balance_loss_mlp": 1.02477694,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 5.277705798532969,
      "language_loss": 0.77932894,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.80139387,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.749868154525757
    },
    {
      "auxiliary_loss_clip": 0.01135836,
      "auxiliary_loss_mlp": 0.01031204,
      "balance_loss_clip": 1.0447433,
      "balance_loss_mlp": 1.02105916,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 1.906305087257039,
      "language_loss": 0.75532812,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77699852,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.8138747215270996
    },
    {
      "auxiliary_loss_clip": 0.01193925,
      "auxiliary_loss_mlp": 0.0103797,
      "balance_loss_clip": 1.05829012,
      "balance_loss_mlp": 1.0289222,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 8.8052390702638,
      "language_loss": 0.79057777,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81289673,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.6940038204193115
    },
    {
      "auxiliary_loss_clip": 0.0119044,
      "auxiliary_loss_mlp": 0.01038681,
      "balance_loss_clip": 1.05548108,
      "balance_loss_mlp": 1.028548,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 14.264893414378985,
      "language_loss": 0.8199048,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84219599,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.699328899383545
    },
    {
      "auxiliary_loss_clip": 0.01190497,
      "auxiliary_loss_mlp": 0.01032858,
      "balance_loss_clip": 1.05558145,
      "balance_loss_mlp": 1.02326715,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 3.332407490815772,
      "language_loss": 0.80932307,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83155656,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.6946098804473877
    },
    {
      "auxiliary_loss_clip": 0.01206479,
      "auxiliary_loss_mlp": 0.00902763,
      "balance_loss_clip": 1.05880189,
      "balance_loss_mlp": 1.00220656,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 2.6417885555479654,
      "language_loss": 0.74129295,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.76238537,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 3.51362943649292
    },
    {
      "auxiliary_loss_clip": 0.01118061,
      "auxiliary_loss_mlp": 0.01005769,
      "balance_loss_clip": 1.03473616,
      "balance_loss_mlp": 1.00361145,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7840993137534291,
      "language_loss": 0.64816976,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66940808,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 3.1417524814605713
    },
    {
      "auxiliary_loss_clip": 0.01177266,
      "auxiliary_loss_mlp": 0.01034909,
      "balance_loss_clip": 1.0517695,
      "balance_loss_mlp": 1.02597427,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.8977909117387097,
      "language_loss": 0.90371597,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.9258377,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 3.5909171104431152
    },
    {
      "auxiliary_loss_clip": 0.01204538,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.05796468,
      "balance_loss_mlp": 1.01757324,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 3.8154793809370515,
      "language_loss": 0.81538737,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83769625,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.5865139961242676
    },
    {
      "auxiliary_loss_clip": 0.01207665,
      "auxiliary_loss_mlp": 0.01037181,
      "balance_loss_clip": 1.05841374,
      "balance_loss_mlp": 1.0277158,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 3.4987220715524106,
      "language_loss": 0.71427238,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.7367208,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.5875864028930664
    },
    {
      "auxiliary_loss_clip": 0.01180971,
      "auxiliary_loss_mlp": 0.01033072,
      "balance_loss_clip": 1.05357742,
      "balance_loss_mlp": 1.02343965,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.9827639165211002,
      "language_loss": 0.77842009,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80056047,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.7086904048919678
    },
    {
      "auxiliary_loss_clip": 0.01171529,
      "auxiliary_loss_mlp": 0.01029058,
      "balance_loss_clip": 1.05008078,
      "balance_loss_mlp": 1.01990867,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.6012339368430037,
      "language_loss": 0.83340192,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85540783,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.699277639389038
    },
    {
      "auxiliary_loss_clip": 0.01187439,
      "auxiliary_loss_mlp": 0.01030885,
      "balance_loss_clip": 1.05660152,
      "balance_loss_mlp": 1.02153325,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.354787281782133,
      "language_loss": 0.8216601,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84384334,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.6911110877990723
    },
    {
      "auxiliary_loss_clip": 0.01194595,
      "auxiliary_loss_mlp": 0.01036894,
      "balance_loss_clip": 1.05649352,
      "balance_loss_mlp": 1.02875805,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 2.142118998414783,
      "language_loss": 0.88069272,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90300763,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 3.5939857959747314
    },
    {
      "auxiliary_loss_clip": 0.01129036,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.04526532,
      "balance_loss_mlp": 1.02271879,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 2.1545995836823613,
      "language_loss": 0.77493215,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79654002,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 2.970980644226074
    },
    {
      "auxiliary_loss_clip": 0.01190794,
      "auxiliary_loss_mlp": 0.01032177,
      "balance_loss_clip": 1.05445528,
      "balance_loss_mlp": 1.023206,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.086027627050994,
      "language_loss": 0.75538903,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77761877,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.6496949195861816
    },
    {
      "auxiliary_loss_clip": 0.01158281,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.05033779,
      "balance_loss_mlp": 1.02020741,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 2.162238898005285,
      "language_loss": 0.70731801,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.72919285,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.7653846740722656
    },
    {
      "auxiliary_loss_clip": 0.01161491,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.05453849,
      "balance_loss_mlp": 1.02224398,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 2.2764563654770913,
      "language_loss": 0.74104434,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76297677,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.7455849647521973
    },
    {
      "auxiliary_loss_clip": 0.01179934,
      "auxiliary_loss_mlp": 0.01029883,
      "balance_loss_clip": 1.0515275,
      "balance_loss_mlp": 1.02107382,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.6196250736031235,
      "language_loss": 0.81467462,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.8367728,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.7155404090881348
    },
    {
      "auxiliary_loss_clip": 0.01074279,
      "auxiliary_loss_mlp": 0.01006576,
      "balance_loss_clip": 1.02649438,
      "balance_loss_mlp": 1.00441861,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9431029915679864,
      "language_loss": 0.60806704,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62887555,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.3036458492279053
    },
    {
      "auxiliary_loss_clip": 0.01191716,
      "auxiliary_loss_mlp": 0.0103405,
      "balance_loss_clip": 1.05744326,
      "balance_loss_mlp": 1.02506757,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.7612099388286215,
      "language_loss": 0.83715487,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85941255,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.650242328643799
    },
    {
      "auxiliary_loss_clip": 0.01181777,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.05477679,
      "balance_loss_mlp": 1.01808286,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 2.5162836203637013,
      "language_loss": 0.75490892,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.7769978,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.6293745040893555
    },
    {
      "auxiliary_loss_clip": 0.01172399,
      "auxiliary_loss_mlp": 0.01032502,
      "balance_loss_clip": 1.05353856,
      "balance_loss_mlp": 1.02320921,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 2.7647999598702344,
      "language_loss": 0.8346082,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85665721,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.757874011993408
    },
    {
      "auxiliary_loss_clip": 0.01205045,
      "auxiliary_loss_mlp": 0.01032903,
      "balance_loss_clip": 1.05783415,
      "balance_loss_mlp": 1.02459443,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 4.049753263210273,
      "language_loss": 0.87635744,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.89873689,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.5642080307006836
    },
    {
      "auxiliary_loss_clip": 0.01182355,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.05351281,
      "balance_loss_mlp": 1.02097034,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.7237662732017314,
      "language_loss": 0.80132473,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.82344353,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.6355514526367188
    },
    {
      "auxiliary_loss_clip": 0.01192657,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.05684578,
      "balance_loss_mlp": 1.01911616,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 2.399327680760126,
      "language_loss": 0.75535226,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77755243,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.687744617462158
    },
    {
      "auxiliary_loss_clip": 0.01200904,
      "auxiliary_loss_mlp": 0.01029134,
      "balance_loss_clip": 1.05735755,
      "balance_loss_mlp": 1.01988971,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.5619328139581365,
      "language_loss": 0.84490216,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86720264,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.5653488636016846
    },
    {
      "auxiliary_loss_clip": 0.01200422,
      "auxiliary_loss_mlp": 0.01033891,
      "balance_loss_clip": 1.06718969,
      "balance_loss_mlp": 1.0239017,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 2.49262365793149,
      "language_loss": 0.89988315,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.92222619,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.6034018993377686
    },
    {
      "auxiliary_loss_clip": 0.01204673,
      "auxiliary_loss_mlp": 0.01028865,
      "balance_loss_clip": 1.06096172,
      "balance_loss_mlp": 1.02096725,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 3.2127125621892834,
      "language_loss": 0.73142362,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75375903,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.669832468032837
    },
    {
      "auxiliary_loss_clip": 0.0117609,
      "auxiliary_loss_mlp": 0.01032037,
      "balance_loss_clip": 1.05583858,
      "balance_loss_mlp": 1.02350211,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 2.2011366305901414,
      "language_loss": 0.92660892,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94869018,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.6990854740142822
    },
    {
      "auxiliary_loss_clip": 0.01181827,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.05728269,
      "balance_loss_mlp": 1.0213778,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.1519814200555984,
      "language_loss": 0.82488984,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84700561,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.691199541091919
    },
    {
      "auxiliary_loss_clip": 0.01189107,
      "auxiliary_loss_mlp": 0.01028832,
      "balance_loss_clip": 1.05640864,
      "balance_loss_mlp": 1.01992679,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 2.2866458209145497,
      "language_loss": 0.87112659,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89330602,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.624025344848633
    },
    {
      "auxiliary_loss_clip": 0.01173915,
      "auxiliary_loss_mlp": 0.01029689,
      "balance_loss_clip": 1.05319619,
      "balance_loss_mlp": 1.02114797,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.172824474461668,
      "language_loss": 0.82725954,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84929562,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 3.5938076972961426
    },
    {
      "auxiliary_loss_clip": 0.01170017,
      "auxiliary_loss_mlp": 0.01030704,
      "balance_loss_clip": 1.04891837,
      "balance_loss_mlp": 1.02139962,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.8234948851514798,
      "language_loss": 0.7852248,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80723202,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 3.705343246459961
    },
    {
      "auxiliary_loss_clip": 0.01146093,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.04614878,
      "balance_loss_mlp": 1.01536298,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 2.1135922728698113,
      "language_loss": 0.86256623,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88425958,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 3.639317750930786
    },
    {
      "auxiliary_loss_clip": 0.01201801,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.05679822,
      "balance_loss_mlp": 1.02185559,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 2.472899000638794,
      "language_loss": 0.8252272,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84754586,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.626451015472412
    },
    {
      "auxiliary_loss_clip": 0.01106132,
      "auxiliary_loss_mlp": 0.01007514,
      "balance_loss_clip": 1.03284872,
      "balance_loss_mlp": 1.00552297,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7923541895345152,
      "language_loss": 0.60259467,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.6237312,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.144862651824951
    },
    {
      "auxiliary_loss_clip": 0.0119183,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.05479765,
      "balance_loss_mlp": 1.02265632,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 1.9503184729285445,
      "language_loss": 0.82955623,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.85179567,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.608314037322998
    },
    {
      "auxiliary_loss_clip": 0.01179436,
      "auxiliary_loss_mlp": 0.01034325,
      "balance_loss_clip": 1.05216479,
      "balance_loss_mlp": 1.02543759,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 1.9467995980330717,
      "language_loss": 0.87862766,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90076524,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.6185483932495117
    },
    {
      "auxiliary_loss_clip": 0.01201364,
      "auxiliary_loss_mlp": 0.0102608,
      "balance_loss_clip": 1.05655229,
      "balance_loss_mlp": 1.01780057,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 3.113909888376127,
      "language_loss": 0.75215966,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77443409,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.603362560272217
    },
    {
      "auxiliary_loss_clip": 0.01161592,
      "auxiliary_loss_mlp": 0.01030425,
      "balance_loss_clip": 1.05508792,
      "balance_loss_mlp": 1.02187157,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 2.131985804349073,
      "language_loss": 0.78858352,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81050372,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 3.6701698303222656
    },
    {
      "auxiliary_loss_clip": 0.01182853,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.05693674,
      "balance_loss_mlp": 1.02458334,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.8892472032571441,
      "language_loss": 0.74063319,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76279426,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.7584033012390137
    },
    {
      "auxiliary_loss_clip": 0.01181941,
      "auxiliary_loss_mlp": 0.01037187,
      "balance_loss_clip": 1.05621839,
      "balance_loss_mlp": 1.02791834,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 3.50205434963957,
      "language_loss": 0.65878844,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.68097973,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.6496496200561523
    },
    {
      "auxiliary_loss_clip": 0.01185193,
      "auxiliary_loss_mlp": 0.01024213,
      "balance_loss_clip": 1.05385256,
      "balance_loss_mlp": 1.01596975,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.533526168991055,
      "language_loss": 0.82351798,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84561205,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.6510305404663086
    },
    {
      "auxiliary_loss_clip": 0.01207487,
      "auxiliary_loss_mlp": 0.01033877,
      "balance_loss_clip": 1.06058359,
      "balance_loss_mlp": 1.02466774,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.2611530277888474,
      "language_loss": 0.91542137,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93783492,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.6712570190429688
    },
    {
      "auxiliary_loss_clip": 0.01194161,
      "auxiliary_loss_mlp": 0.01023449,
      "balance_loss_clip": 1.05703187,
      "balance_loss_mlp": 1.01531863,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 2.2486091116487343,
      "language_loss": 0.80427247,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82644856,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.6039838790893555
    },
    {
      "auxiliary_loss_clip": 0.01193763,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.05732512,
      "balance_loss_mlp": 1.01685226,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.270120197507109,
      "language_loss": 0.75081176,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77300584,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.747476100921631
    },
    {
      "auxiliary_loss_clip": 0.01108204,
      "auxiliary_loss_mlp": 0.00893999,
      "balance_loss_clip": 1.0345583,
      "balance_loss_mlp": 1.00118816,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8463611236161598,
      "language_loss": 0.69713688,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71715891,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.238215684890747
    },
    {
      "auxiliary_loss_clip": 0.01157354,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.05127454,
      "balance_loss_mlp": 1.01785743,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.57490193966391,
      "language_loss": 0.74094248,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76279044,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.7807693481445312
    },
    {
      "auxiliary_loss_clip": 0.01194507,
      "auxiliary_loss_mlp": 0.01028447,
      "balance_loss_clip": 1.05699146,
      "balance_loss_mlp": 1.02019167,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.367646767733336,
      "language_loss": 0.65182424,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67405379,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.643747329711914
    },
    {
      "auxiliary_loss_clip": 0.01193787,
      "auxiliary_loss_mlp": 0.0102746,
      "balance_loss_clip": 1.05981374,
      "balance_loss_mlp": 1.01883543,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 2.347682809614797,
      "language_loss": 0.73885816,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.76107061,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.675213575363159
    },
    {
      "auxiliary_loss_clip": 0.01181159,
      "auxiliary_loss_mlp": 0.01029859,
      "balance_loss_clip": 1.05503988,
      "balance_loss_mlp": 1.02107906,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 3.11792621763732,
      "language_loss": 0.7153157,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73742592,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.638545036315918
    },
    {
      "auxiliary_loss_clip": 0.01186663,
      "auxiliary_loss_mlp": 0.01029125,
      "balance_loss_clip": 1.05609727,
      "balance_loss_mlp": 1.02073312,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.962697436244367,
      "language_loss": 0.78455323,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80671108,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.7861876487731934
    },
    {
      "auxiliary_loss_clip": 0.01195727,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.06333137,
      "balance_loss_mlp": 1.01958489,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 2.5826857248204287,
      "language_loss": 0.78175282,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80399233,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.743433952331543
    },
    {
      "auxiliary_loss_clip": 0.01179626,
      "auxiliary_loss_mlp": 0.01030053,
      "balance_loss_clip": 1.05440593,
      "balance_loss_mlp": 1.02097523,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 2.244176330585776,
      "language_loss": 0.87761796,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.89971471,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.632758617401123
    },
    {
      "auxiliary_loss_clip": 0.01199662,
      "auxiliary_loss_mlp": 0.01028589,
      "balance_loss_clip": 1.05895174,
      "balance_loss_mlp": 1.02047682,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.860497668535193,
      "language_loss": 0.78700173,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80928427,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.682926654815674
    },
    {
      "auxiliary_loss_clip": 0.01188791,
      "auxiliary_loss_mlp": 0.01030044,
      "balance_loss_clip": 1.05274057,
      "balance_loss_mlp": 1.02087677,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 2.13168160581868,
      "language_loss": 0.69403195,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71622032,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.764505386352539
    },
    {
      "auxiliary_loss_clip": 0.01179357,
      "auxiliary_loss_mlp": 0.00902236,
      "balance_loss_clip": 1.0544312,
      "balance_loss_mlp": 1.00194728,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 2.1910147716570756,
      "language_loss": 0.82926959,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.8500855,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.7412362098693848
    },
    {
      "auxiliary_loss_clip": 0.01203518,
      "auxiliary_loss_mlp": 0.01030158,
      "balance_loss_clip": 1.06004238,
      "balance_loss_mlp": 1.02172995,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.60831823704161,
      "language_loss": 0.85110402,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.8734408,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 3.5713980197906494
    },
    {
      "auxiliary_loss_clip": 0.01169577,
      "auxiliary_loss_mlp": 0.01021355,
      "balance_loss_clip": 1.05087125,
      "balance_loss_mlp": 1.01293337,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.0587735723150007,
      "language_loss": 0.79577625,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.8176856,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 3.6935787200927734
    },
    {
      "auxiliary_loss_clip": 0.01188633,
      "auxiliary_loss_mlp": 0.01026255,
      "balance_loss_clip": 1.05133641,
      "balance_loss_mlp": 1.01774907,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.74778161101742,
      "language_loss": 0.68366796,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.70581686,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.6798465251922607
    },
    {
      "auxiliary_loss_clip": 0.01161908,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.04890084,
      "balance_loss_mlp": 1.01923919,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 4.061911524269864,
      "language_loss": 0.70551127,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72740483,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 3.663431167602539
    },
    {
      "auxiliary_loss_clip": 0.01156086,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.04600048,
      "balance_loss_mlp": 1.02162743,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.1393367303156308,
      "language_loss": 0.74052405,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76239216,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.792461395263672
    },
    {
      "auxiliary_loss_clip": 0.01187578,
      "auxiliary_loss_mlp": 0.01028458,
      "balance_loss_clip": 1.05298781,
      "balance_loss_mlp": 1.01940382,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 2.734066001629722,
      "language_loss": 0.81379688,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83595729,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.686835289001465
    },
    {
      "auxiliary_loss_clip": 0.01175685,
      "auxiliary_loss_mlp": 0.01032723,
      "balance_loss_clip": 1.05449104,
      "balance_loss_mlp": 1.02341866,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.0672680873916742,
      "language_loss": 0.71725863,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.73934269,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.798062562942505
    },
    {
      "auxiliary_loss_clip": 0.01189111,
      "auxiliary_loss_mlp": 0.01021801,
      "balance_loss_clip": 1.0571084,
      "balance_loss_mlp": 1.01401639,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 2.0193000985641274,
      "language_loss": 0.8095603,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.83166939,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.6432602405548096
    },
    {
      "auxiliary_loss_clip": 0.01119553,
      "auxiliary_loss_mlp": 0.01003959,
      "balance_loss_clip": 1.03428698,
      "balance_loss_mlp": 1.00155067,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7820556447412685,
      "language_loss": 0.58087808,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60211319,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 4.066727876663208
    },
    {
      "auxiliary_loss_clip": 0.01193893,
      "auxiliary_loss_mlp": 0.01027092,
      "balance_loss_clip": 1.05463648,
      "balance_loss_mlp": 1.01840806,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 3.574912641459626,
      "language_loss": 0.94871128,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.97092116,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.623368740081787
    },
    {
      "auxiliary_loss_clip": 0.01184682,
      "auxiliary_loss_mlp": 0.01032376,
      "balance_loss_clip": 1.05393791,
      "balance_loss_mlp": 1.02281582,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.5055059171590353,
      "language_loss": 0.78562081,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80779141,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.619800567626953
    },
    {
      "auxiliary_loss_clip": 0.01159415,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.05368638,
      "balance_loss_mlp": 1.01920867,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.893595260148862,
      "language_loss": 0.80316877,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82503939,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.723954200744629
    },
    {
      "auxiliary_loss_clip": 0.01184431,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.05649114,
      "balance_loss_mlp": 1.02229118,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.2667415454073674,
      "language_loss": 0.70111704,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.72327268,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.705963134765625
    },
    {
      "auxiliary_loss_clip": 0.01180196,
      "auxiliary_loss_mlp": 0.01034075,
      "balance_loss_clip": 1.05516458,
      "balance_loss_mlp": 1.02583742,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 2.5226444621078836,
      "language_loss": 0.9001689,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.9223116,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.638723134994507
    },
    {
      "auxiliary_loss_clip": 0.01168135,
      "auxiliary_loss_mlp": 0.00902894,
      "balance_loss_clip": 1.05062604,
      "balance_loss_mlp": 1.00194907,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 2.1229650102699544,
      "language_loss": 0.83312112,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85383141,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.7604458332061768
    },
    {
      "auxiliary_loss_clip": 0.01190139,
      "auxiliary_loss_mlp": 0.01028162,
      "balance_loss_clip": 1.05288267,
      "balance_loss_mlp": 1.01994264,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 1.7894562619776833,
      "language_loss": 0.74191403,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76409709,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.714233636856079
    },
    {
      "auxiliary_loss_clip": 0.01171882,
      "auxiliary_loss_mlp": 0.01023214,
      "balance_loss_clip": 1.05125403,
      "balance_loss_mlp": 1.01475573,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 2.143535017755232,
      "language_loss": 0.82955682,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85150778,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.748563528060913
    },
    {
      "auxiliary_loss_clip": 0.01192839,
      "auxiliary_loss_mlp": 0.0103193,
      "balance_loss_clip": 1.05310106,
      "balance_loss_mlp": 1.02373433,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 2.9490979697723585,
      "language_loss": 0.74958116,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77182889,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.610948085784912
    },
    {
      "auxiliary_loss_clip": 0.01177131,
      "auxiliary_loss_mlp": 0.01031912,
      "balance_loss_clip": 1.05015457,
      "balance_loss_mlp": 1.02288198,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 2.6479010022563516,
      "language_loss": 0.86952096,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89161134,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.712447166442871
    },
    {
      "auxiliary_loss_clip": 0.01206934,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.05897272,
      "balance_loss_mlp": 1.02157998,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.0308051381427465,
      "language_loss": 0.80613571,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82850766,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.588550567626953
    },
    {
      "auxiliary_loss_clip": 0.01112615,
      "auxiliary_loss_mlp": 0.01001522,
      "balance_loss_clip": 1.04884195,
      "balance_loss_mlp": 0.99979365,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.745748088782805,
      "language_loss": 0.60643196,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62757337,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.3280701637268066
    },
    {
      "auxiliary_loss_clip": 0.01179795,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04994643,
      "balance_loss_mlp": 1.01875782,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 2.3497389795117427,
      "language_loss": 0.72141415,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74348921,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.72107195854187
    },
    {
      "auxiliary_loss_clip": 0.0115158,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.04611874,
      "balance_loss_mlp": 1.01832175,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.9794500674533437,
      "language_loss": 0.86377311,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88556027,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.755524158477783
    },
    {
      "auxiliary_loss_clip": 0.01176171,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.05286789,
      "balance_loss_mlp": 1.0164808,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 3.33923434542825,
      "language_loss": 0.80061764,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82262778,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.6921706199645996
    },
    {
      "auxiliary_loss_clip": 0.01187414,
      "auxiliary_loss_mlp": 0.01030147,
      "balance_loss_clip": 1.05407023,
      "balance_loss_mlp": 1.02145076,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 1.8782097027929014,
      "language_loss": 0.80412668,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.82630229,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.563971757888794
    },
    {
      "auxiliary_loss_clip": 0.01192634,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.0561564,
      "balance_loss_mlp": 1.01860023,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.087672810613818,
      "language_loss": 0.71854085,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.74074495,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.677053451538086
    },
    {
      "auxiliary_loss_clip": 0.01181849,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.05328918,
      "balance_loss_mlp": 1.02406049,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.4299004125378767,
      "language_loss": 0.66260308,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68474919,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 3.6657159328460693
    },
    {
      "auxiliary_loss_clip": 0.01197403,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.05963087,
      "balance_loss_mlp": 1.01814115,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.8985375975027567,
      "language_loss": 0.73364854,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.7558977,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.7287678718566895
    },
    {
      "auxiliary_loss_clip": 0.01117865,
      "auxiliary_loss_mlp": 0.01010303,
      "balance_loss_clip": 1.03743505,
      "balance_loss_mlp": 1.00799072,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.219787410930063,
      "language_loss": 0.68319255,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70447427,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 4.041267395019531
    },
    {
      "auxiliary_loss_clip": 0.01178942,
      "auxiliary_loss_mlp": 0.01030907,
      "balance_loss_clip": 1.0506537,
      "balance_loss_mlp": 1.02185285,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 6.823785997380146,
      "language_loss": 0.76870656,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.79080498,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 3.6278762817382812
    },
    {
      "auxiliary_loss_clip": 0.01144337,
      "auxiliary_loss_mlp": 0.00903514,
      "balance_loss_clip": 1.04907584,
      "balance_loss_mlp": 1.00190711,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 1.6985430543638462,
      "language_loss": 0.81546992,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83594847,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.8466877937316895
    },
    {
      "auxiliary_loss_clip": 0.01182087,
      "auxiliary_loss_mlp": 0.01041412,
      "balance_loss_clip": 1.05370641,
      "balance_loss_mlp": 1.03164291,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.162637675206982,
      "language_loss": 0.79103512,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81327009,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.6676077842712402
    },
    {
      "auxiliary_loss_clip": 0.01178964,
      "auxiliary_loss_mlp": 0.01025197,
      "balance_loss_clip": 1.05017364,
      "balance_loss_mlp": 1.01605415,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.263174864281947,
      "language_loss": 0.75668991,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.77873158,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.6259448528289795
    },
    {
      "auxiliary_loss_clip": 0.01193383,
      "auxiliary_loss_mlp": 0.01032701,
      "balance_loss_clip": 1.05471706,
      "balance_loss_mlp": 1.02358782,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.7592344451967374,
      "language_loss": 0.80759525,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.82985604,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.6596968173980713
    },
    {
      "auxiliary_loss_clip": 0.01165195,
      "auxiliary_loss_mlp": 0.00902987,
      "balance_loss_clip": 1.04855704,
      "balance_loss_mlp": 1.00187039,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.6838445060105935,
      "language_loss": 0.81432074,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.8350026,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 3.6689133644104004
    },
    {
      "auxiliary_loss_clip": 0.01205969,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 1.05830193,
      "balance_loss_mlp": 1.02193403,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 3.503634657215282,
      "language_loss": 0.90509105,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92745793,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.606987953186035
    },
    {
      "auxiliary_loss_clip": 0.01164351,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 1.04586613,
      "balance_loss_mlp": 1.021896,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.9775085132909276,
      "language_loss": 0.65817928,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68013787,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.9100871086120605
    },
    {
      "auxiliary_loss_clip": 0.01156224,
      "auxiliary_loss_mlp": 0.01022358,
      "balance_loss_clip": 1.04836273,
      "balance_loss_mlp": 1.01362002,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 2.2950414585652084,
      "language_loss": 0.9017694,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.9235552,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.781660318374634
    },
    {
      "auxiliary_loss_clip": 0.01186354,
      "auxiliary_loss_mlp": 0.01030665,
      "balance_loss_clip": 1.05215037,
      "balance_loss_mlp": 1.02084792,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.7870881440516024,
      "language_loss": 0.80225742,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.8244276,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.6534276008605957
    },
    {
      "auxiliary_loss_clip": 0.01178716,
      "auxiliary_loss_mlp": 0.01037588,
      "balance_loss_clip": 1.05679965,
      "balance_loss_mlp": 1.02864122,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.2141130206447075,
      "language_loss": 0.93256092,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.95472395,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.6919050216674805
    },
    {
      "auxiliary_loss_clip": 0.0118709,
      "auxiliary_loss_mlp": 0.01034385,
      "balance_loss_clip": 1.05458987,
      "balance_loss_mlp": 1.02506912,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 2.4745666650030995,
      "language_loss": 0.80782723,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.830042,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.6371841430664062
    },
    {
      "auxiliary_loss_clip": 0.01112565,
      "auxiliary_loss_mlp": 0.01000787,
      "balance_loss_clip": 1.03238821,
      "balance_loss_mlp": 0.99856967,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.8542449980908848,
      "language_loss": 0.56957436,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.5907079,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.2132437229156494
    },
    {
      "auxiliary_loss_clip": 0.01166344,
      "auxiliary_loss_mlp": 0.01032894,
      "balance_loss_clip": 1.05224562,
      "balance_loss_mlp": 1.02325583,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.9420131268434018,
      "language_loss": 0.84987509,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87186742,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.7520298957824707
    },
    {
      "auxiliary_loss_clip": 0.0117312,
      "auxiliary_loss_mlp": 0.01028059,
      "balance_loss_clip": 1.05077732,
      "balance_loss_mlp": 1.01930285,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 3.5418776062417066,
      "language_loss": 0.83023918,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85225093,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.6210038661956787
    },
    {
      "auxiliary_loss_clip": 0.01162395,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.0501411,
      "balance_loss_mlp": 1.01951253,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 2.172421079550831,
      "language_loss": 0.81005633,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.83196521,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.676173686981201
    },
    {
      "auxiliary_loss_clip": 0.01198633,
      "auxiliary_loss_mlp": 0.01033293,
      "balance_loss_clip": 1.0562607,
      "balance_loss_mlp": 1.02326715,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 1.9418029107643529,
      "language_loss": 0.85127413,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87359339,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.6112046241760254
    },
    {
      "auxiliary_loss_clip": 0.01190619,
      "auxiliary_loss_mlp": 0.01033162,
      "balance_loss_clip": 1.05455863,
      "balance_loss_mlp": 1.02388191,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.2557112620337523,
      "language_loss": 0.62284118,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64507896,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.6781959533691406
    },
    {
      "auxiliary_loss_clip": 0.01190717,
      "auxiliary_loss_mlp": 0.01036549,
      "balance_loss_clip": 1.05723238,
      "balance_loss_mlp": 1.02729273,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 4.1336138779304425,
      "language_loss": 0.78612423,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80839694,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.7308530807495117
    },
    {
      "auxiliary_loss_clip": 0.01177239,
      "auxiliary_loss_mlp": 0.01033875,
      "balance_loss_clip": 1.05281579,
      "balance_loss_mlp": 1.02486885,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 3.3763217674591988,
      "language_loss": 0.86387402,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88598514,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.943653106689453
    },
    {
      "auxiliary_loss_clip": 0.01169005,
      "auxiliary_loss_mlp": 0.0103557,
      "balance_loss_clip": 1.04937029,
      "balance_loss_mlp": 1.02557421,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 3.002432123933849,
      "language_loss": 0.71028066,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73232645,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.7549564838409424
    },
    {
      "auxiliary_loss_clip": 0.01197052,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.05927014,
      "balance_loss_mlp": 1.0262351,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 2.3077919968177203,
      "language_loss": 0.70103979,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72336769,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.6167635917663574
    },
    {
      "auxiliary_loss_clip": 0.01183599,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.05281222,
      "balance_loss_mlp": 1.01886749,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 2.1260211964280775,
      "language_loss": 0.79310715,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81522757,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.704542636871338
    },
    {
      "auxiliary_loss_clip": 0.01172495,
      "auxiliary_loss_mlp": 0.01030316,
      "balance_loss_clip": 1.0543319,
      "balance_loss_mlp": 1.02001071,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.2452404629047327,
      "language_loss": 0.8459214,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86794949,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.739039182662964
    },
    {
      "auxiliary_loss_clip": 0.01205275,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.05805016,
      "balance_loss_mlp": 1.02116656,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 2.739831444154539,
      "language_loss": 0.89824939,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.92060375,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 3.6131365299224854
    },
    {
      "auxiliary_loss_clip": 0.01191609,
      "auxiliary_loss_mlp": 0.01032668,
      "balance_loss_clip": 1.05535078,
      "balance_loss_mlp": 1.02298832,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 2.6197034862887625,
      "language_loss": 0.8545211,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.87676388,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.6824710369110107
    },
    {
      "auxiliary_loss_clip": 0.01170248,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.05197692,
      "balance_loss_mlp": 1.02296996,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 1.8943748573636061,
      "language_loss": 0.7722348,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.7942571,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 3.6845147609710693
    },
    {
      "auxiliary_loss_clip": 0.01208555,
      "auxiliary_loss_mlp": 0.01026733,
      "balance_loss_clip": 1.06095016,
      "balance_loss_mlp": 1.01716065,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.248482056251932,
      "language_loss": 0.77140135,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79375422,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 3.5238304138183594
    },
    {
      "auxiliary_loss_clip": 0.01167585,
      "auxiliary_loss_mlp": 0.01034779,
      "balance_loss_clip": 1.05306911,
      "balance_loss_mlp": 1.02543914,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.9475855208802793,
      "language_loss": 0.77774197,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.79976559,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.772202491760254
    },
    {
      "auxiliary_loss_clip": 0.01207631,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.06032348,
      "balance_loss_mlp": 1.02018571,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 2.655668619087074,
      "language_loss": 0.76664507,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.78901565,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.6772449016571045
    },
    {
      "auxiliary_loss_clip": 0.01184237,
      "auxiliary_loss_mlp": 0.01034175,
      "balance_loss_clip": 1.05488336,
      "balance_loss_mlp": 1.0252763,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 3.6391597818114345,
      "language_loss": 0.79568064,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81786478,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.7479310035705566
    },
    {
      "auxiliary_loss_clip": 0.01175183,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.0523169,
      "balance_loss_mlp": 1.01669836,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.7543549194809867,
      "language_loss": 0.7396456,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.7616635,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 3.5907530784606934
    },
    {
      "auxiliary_loss_clip": 0.01181622,
      "auxiliary_loss_mlp": 0.01030148,
      "balance_loss_clip": 1.05862927,
      "balance_loss_mlp": 1.02105236,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.874187204596418,
      "language_loss": 0.72820276,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75032049,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.707933187484741
    },
    {
      "auxiliary_loss_clip": 0.01188304,
      "auxiliary_loss_mlp": 0.01032705,
      "balance_loss_clip": 1.05709219,
      "balance_loss_mlp": 1.02232182,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.6626569578039962,
      "language_loss": 0.83631486,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85852492,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.6184306144714355
    },
    {
      "auxiliary_loss_clip": 0.0118001,
      "auxiliary_loss_mlp": 0.0103137,
      "balance_loss_clip": 1.05209112,
      "balance_loss_mlp": 1.02173197,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 4.491100767237787,
      "language_loss": 0.53612322,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.55823708,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.764880657196045
    },
    {
      "auxiliary_loss_clip": 0.01198061,
      "auxiliary_loss_mlp": 0.01030733,
      "balance_loss_clip": 1.0603863,
      "balance_loss_mlp": 1.02129173,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.5025676639493057,
      "language_loss": 0.75432962,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77661753,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.599656581878662
    },
    {
      "auxiliary_loss_clip": 0.0120539,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.06090641,
      "balance_loss_mlp": 1.02272558,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 2.2935291731519367,
      "language_loss": 0.636994,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.65937555,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.6189401149749756
    },
    {
      "auxiliary_loss_clip": 0.01190241,
      "auxiliary_loss_mlp": 0.01039294,
      "balance_loss_clip": 1.05695248,
      "balance_loss_mlp": 1.03021586,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 3.3058887003902604,
      "language_loss": 0.85262507,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87492043,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.7388482093811035
    },
    {
      "auxiliary_loss_clip": 0.01185977,
      "auxiliary_loss_mlp": 0.01033477,
      "balance_loss_clip": 1.05584979,
      "balance_loss_mlp": 1.02460229,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 1.8775460891887112,
      "language_loss": 0.88179517,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.90398967,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.693255662918091
    },
    {
      "auxiliary_loss_clip": 0.01163586,
      "auxiliary_loss_mlp": 0.01036127,
      "balance_loss_clip": 1.05473804,
      "balance_loss_mlp": 1.02689385,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 2.108570025968018,
      "language_loss": 0.92669851,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94869566,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.802720785140991
    },
    {
      "auxiliary_loss_clip": 0.01184468,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.05458784,
      "balance_loss_mlp": 1.01960993,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.319552206535058,
      "language_loss": 0.88227463,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90441549,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.6904830932617188
    },
    {
      "auxiliary_loss_clip": 0.01200616,
      "auxiliary_loss_mlp": 0.01033707,
      "balance_loss_clip": 1.06223667,
      "balance_loss_mlp": 1.02319288,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 2.0906327078891507,
      "language_loss": 0.83756971,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85991299,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.76644229888916
    },
    {
      "auxiliary_loss_clip": 0.01211135,
      "auxiliary_loss_mlp": 0.0103334,
      "balance_loss_clip": 1.06139207,
      "balance_loss_mlp": 1.02174687,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 1.9644870508762502,
      "language_loss": 0.83097458,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.8534193,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.642266273498535
    },
    {
      "auxiliary_loss_clip": 0.01189287,
      "auxiliary_loss_mlp": 0.01036769,
      "balance_loss_clip": 1.0561409,
      "balance_loss_mlp": 1.02626073,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.6276888840512418,
      "language_loss": 0.78090644,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80316699,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.7121732234954834
    },
    {
      "auxiliary_loss_clip": 0.01207663,
      "auxiliary_loss_mlp": 0.01038555,
      "balance_loss_clip": 1.06156778,
      "balance_loss_mlp": 1.02874994,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 2.30351667270141,
      "language_loss": 0.826855,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.84931719,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.6984028816223145
    },
    {
      "auxiliary_loss_clip": 0.01186573,
      "auxiliary_loss_mlp": 0.01036624,
      "balance_loss_clip": 1.05577087,
      "balance_loss_mlp": 1.02579379,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 2.0675232090866844,
      "language_loss": 0.84647965,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86871159,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.698622703552246
    },
    {
      "auxiliary_loss_clip": 0.01199651,
      "auxiliary_loss_mlp": 0.01032773,
      "balance_loss_clip": 1.06136799,
      "balance_loss_mlp": 1.02396345,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.8351175359044687,
      "language_loss": 0.83660591,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.85893023,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.6942968368530273
    },
    {
      "auxiliary_loss_clip": 0.01190096,
      "auxiliary_loss_mlp": 0.01037966,
      "balance_loss_clip": 1.05753601,
      "balance_loss_mlp": 1.0281198,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 1.870235921354861,
      "language_loss": 0.74898273,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77126336,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.774646759033203
    },
    {
      "auxiliary_loss_clip": 0.01179656,
      "auxiliary_loss_mlp": 0.01033074,
      "balance_loss_clip": 1.06025136,
      "balance_loss_mlp": 1.02363873,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.1399528785935003,
      "language_loss": 0.83157557,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85370284,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.742445707321167
    },
    {
      "auxiliary_loss_clip": 0.01193194,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.05548429,
      "balance_loss_mlp": 1.02268636,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.771252089756817,
      "language_loss": 0.68512166,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70737672,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.6840412616729736
    },
    {
      "auxiliary_loss_clip": 0.01204428,
      "auxiliary_loss_mlp": 0.010325,
      "balance_loss_clip": 1.06137013,
      "balance_loss_mlp": 1.02268887,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 2.0867293364481374,
      "language_loss": 0.71623892,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73860818,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 3.821498155593872
    },
    {
      "auxiliary_loss_clip": 0.01194593,
      "auxiliary_loss_mlp": 0.01037103,
      "balance_loss_clip": 1.05856609,
      "balance_loss_mlp": 1.02697599,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 3.4453674164596304,
      "language_loss": 0.85847819,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.88079512,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.898423433303833
    },
    {
      "auxiliary_loss_clip": 0.01195369,
      "auxiliary_loss_mlp": 0.00903853,
      "balance_loss_clip": 1.05924249,
      "balance_loss_mlp": 1.00174785,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.0060692359397576,
      "language_loss": 0.67743254,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.6984247,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 3.8497304916381836
    },
    {
      "auxiliary_loss_clip": 0.01188635,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.05671239,
      "balance_loss_mlp": 1.02068973,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 6.104188536310248,
      "language_loss": 0.82557845,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.8477633,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 3.592562437057495
    },
    {
      "auxiliary_loss_clip": 0.0121154,
      "auxiliary_loss_mlp": 0.01028646,
      "balance_loss_clip": 1.06349325,
      "balance_loss_mlp": 1.01937759,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 2.538652622708339,
      "language_loss": 0.75882906,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.78123093,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.8570258617401123
    },
    {
      "auxiliary_loss_clip": 0.01179358,
      "auxiliary_loss_mlp": 0.01035756,
      "balance_loss_clip": 1.05582809,
      "balance_loss_mlp": 1.02541494,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.19686517782862,
      "language_loss": 0.71836042,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74051154,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.8244478702545166
    },
    {
      "auxiliary_loss_clip": 0.01181075,
      "auxiliary_loss_mlp": 0.01028018,
      "balance_loss_clip": 1.0535779,
      "balance_loss_mlp": 1.01913667,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 2.273235436100328,
      "language_loss": 0.91160524,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93369615,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.7379510402679443
    },
    {
      "auxiliary_loss_clip": 0.01161114,
      "auxiliary_loss_mlp": 0.01033156,
      "balance_loss_clip": 1.05705214,
      "balance_loss_mlp": 1.02377462,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.50839501407653,
      "language_loss": 0.77378821,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.79573089,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.8164334297180176
    },
    {
      "auxiliary_loss_clip": 0.01177179,
      "auxiliary_loss_mlp": 0.01031324,
      "balance_loss_clip": 1.0572722,
      "balance_loss_mlp": 1.02216923,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.8943155829919345,
      "language_loss": 0.83448881,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85657382,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 3.6594326496124268
    },
    {
      "auxiliary_loss_clip": 0.01214371,
      "auxiliary_loss_mlp": 0.00903706,
      "balance_loss_clip": 1.06652284,
      "balance_loss_mlp": 1.00184011,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.121710947757409,
      "language_loss": 0.82410139,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84528214,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.5845329761505127
    },
    {
      "auxiliary_loss_clip": 0.01194349,
      "auxiliary_loss_mlp": 0.01029816,
      "balance_loss_clip": 1.06105065,
      "balance_loss_mlp": 1.02041042,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 4.537957102817773,
      "language_loss": 0.76011807,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78235972,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.848383903503418
    },
    {
      "auxiliary_loss_clip": 0.0120158,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.05961335,
      "balance_loss_mlp": 1.01762736,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.7830187239421056,
      "language_loss": 0.8634603,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88573992,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.6212289333343506
    },
    {
      "auxiliary_loss_clip": 0.01200835,
      "auxiliary_loss_mlp": 0.01032741,
      "balance_loss_clip": 1.06341314,
      "balance_loss_mlp": 1.02318645,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.611906512292988,
      "language_loss": 0.81044137,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83277714,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.6432855129241943
    },
    {
      "auxiliary_loss_clip": 0.01209844,
      "auxiliary_loss_mlp": 0.01036774,
      "balance_loss_clip": 1.06205058,
      "balance_loss_mlp": 1.02693939,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 3.093408125277229,
      "language_loss": 0.68751204,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.70997822,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.7715094089508057
    },
    {
      "auxiliary_loss_clip": 0.01171018,
      "auxiliary_loss_mlp": 0.0103132,
      "balance_loss_clip": 1.05442333,
      "balance_loss_mlp": 1.02203393,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 4.782992959297968,
      "language_loss": 0.82972836,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.85175169,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.860612630844116
    },
    {
      "auxiliary_loss_clip": 0.01113135,
      "auxiliary_loss_mlp": 0.01003378,
      "balance_loss_clip": 1.04321384,
      "balance_loss_mlp": 1.00133944,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.830614466107208,
      "language_loss": 0.64668989,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66785502,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.1680238246917725
    },
    {
      "auxiliary_loss_clip": 0.01201373,
      "auxiliary_loss_mlp": 0.01035376,
      "balance_loss_clip": 1.05835366,
      "balance_loss_mlp": 1.025738,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 2.5170770232089508,
      "language_loss": 0.86752552,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.88989305,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.629932165145874
    },
    {
      "auxiliary_loss_clip": 0.01207692,
      "auxiliary_loss_mlp": 0.01034638,
      "balance_loss_clip": 1.06153584,
      "balance_loss_mlp": 1.02575731,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 2.094243162797231,
      "language_loss": 0.75262344,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77504683,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.728893756866455
    },
    {
      "auxiliary_loss_clip": 0.01202207,
      "auxiliary_loss_mlp": 0.01037948,
      "balance_loss_clip": 1.06058741,
      "balance_loss_mlp": 1.02841735,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 1.641913393911628,
      "language_loss": 0.73774642,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.76014799,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.667872667312622
    },
    {
      "auxiliary_loss_clip": 0.01191083,
      "auxiliary_loss_mlp": 0.01035109,
      "balance_loss_clip": 1.05804181,
      "balance_loss_mlp": 1.02498817,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 1.7913969137044115,
      "language_loss": 0.81627399,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.8385359,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.7336392402648926
    },
    {
      "auxiliary_loss_clip": 0.01212802,
      "auxiliary_loss_mlp": 0.00903243,
      "balance_loss_clip": 1.06468725,
      "balance_loss_mlp": 1.00210822,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.921871849464465,
      "language_loss": 0.83525014,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85641062,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.6342363357543945
    },
    {
      "auxiliary_loss_clip": 0.01183582,
      "auxiliary_loss_mlp": 0.01029212,
      "balance_loss_clip": 1.05730557,
      "balance_loss_mlp": 1.01945448,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.6601126188831934,
      "language_loss": 0.89873135,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.92085934,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.6599767208099365
    },
    {
      "auxiliary_loss_clip": 0.01193981,
      "auxiliary_loss_mlp": 0.01034853,
      "balance_loss_clip": 1.06223774,
      "balance_loss_mlp": 1.02539945,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 1.9002622831876281,
      "language_loss": 0.71820581,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74049413,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.8819420337677
    },
    {
      "auxiliary_loss_clip": 0.01150446,
      "auxiliary_loss_mlp": 0.0103349,
      "balance_loss_clip": 1.05131555,
      "balance_loss_mlp": 1.02406073,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.8349757184096867,
      "language_loss": 0.80998337,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83182275,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.760864019393921
    },
    {
      "auxiliary_loss_clip": 0.01208255,
      "auxiliary_loss_mlp": 0.01029791,
      "balance_loss_clip": 1.06444883,
      "balance_loss_mlp": 1.02001011,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.5256200968521352,
      "language_loss": 0.82024378,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.84262419,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.7026171684265137
    },
    {
      "auxiliary_loss_clip": 0.01198511,
      "auxiliary_loss_mlp": 0.01033387,
      "balance_loss_clip": 1.06198406,
      "balance_loss_mlp": 1.02450013,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.992844226818264,
      "language_loss": 0.92339033,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.94570935,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.6583995819091797
    },
    {
      "auxiliary_loss_clip": 0.01210906,
      "auxiliary_loss_mlp": 0.01035075,
      "balance_loss_clip": 1.06233835,
      "balance_loss_mlp": 1.02505493,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 2.072432807982572,
      "language_loss": 0.7068795,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.7293393,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.636392116546631
    },
    {
      "auxiliary_loss_clip": 0.01198354,
      "auxiliary_loss_mlp": 0.01037001,
      "balance_loss_clip": 1.06212366,
      "balance_loss_mlp": 1.02810252,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 2.008072726661362,
      "language_loss": 0.76047099,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.78282452,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 3.6039910316467285
    },
    {
      "auxiliary_loss_clip": 0.0121331,
      "auxiliary_loss_mlp": 0.01030538,
      "balance_loss_clip": 1.06307387,
      "balance_loss_mlp": 1.02107859,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.4765116275777053,
      "language_loss": 0.83553219,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85797065,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.572866678237915
    },
    {
      "auxiliary_loss_clip": 0.01202816,
      "auxiliary_loss_mlp": 0.01027746,
      "balance_loss_clip": 1.05915534,
      "balance_loss_mlp": 1.01868629,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 1.7381220817710847,
      "language_loss": 0.74617994,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76848555,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 3.553356885910034
    },
    {
      "auxiliary_loss_clip": 0.01159162,
      "auxiliary_loss_mlp": 0.01030409,
      "balance_loss_clip": 1.05421937,
      "balance_loss_mlp": 1.02010894,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.2315062077886285,
      "language_loss": 0.62959915,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65149486,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 3.7205684185028076
    },
    {
      "auxiliary_loss_clip": 0.01168247,
      "auxiliary_loss_mlp": 0.01036135,
      "balance_loss_clip": 1.05927455,
      "balance_loss_mlp": 1.02716446,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 2.0587930063212143,
      "language_loss": 0.84313345,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86517727,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.7483456134796143
    },
    {
      "auxiliary_loss_clip": 0.01182063,
      "auxiliary_loss_mlp": 0.01036717,
      "balance_loss_clip": 1.06167519,
      "balance_loss_mlp": 1.02668524,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.7050077607123595,
      "language_loss": 0.83293414,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85512197,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.7853519916534424
    },
    {
      "auxiliary_loss_clip": 0.01175523,
      "auxiliary_loss_mlp": 0.01035433,
      "balance_loss_clip": 1.05643034,
      "balance_loss_mlp": 1.02580118,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.7062978241053703,
      "language_loss": 0.83986175,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86197132,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.6530370712280273
    },
    {
      "auxiliary_loss_clip": 0.01200202,
      "auxiliary_loss_mlp": 0.01030672,
      "balance_loss_clip": 1.0606668,
      "balance_loss_mlp": 1.02092648,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.2261142871772797,
      "language_loss": 0.75681269,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.77912146,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 3.6744284629821777
    },
    {
      "auxiliary_loss_clip": 0.01179521,
      "auxiliary_loss_mlp": 0.01033885,
      "balance_loss_clip": 1.05757189,
      "balance_loss_mlp": 1.02455688,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 1.898309578576087,
      "language_loss": 0.84548509,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86761916,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.7103521823883057
    },
    {
      "auxiliary_loss_clip": 0.01195243,
      "auxiliary_loss_mlp": 0.01033225,
      "balance_loss_clip": 1.0591104,
      "balance_loss_mlp": 1.02360487,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 4.271050964338548,
      "language_loss": 0.76282918,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78511381,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.6366119384765625
    },
    {
      "auxiliary_loss_clip": 0.01185477,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.05771565,
      "balance_loss_mlp": 1.02100825,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.755071196251022,
      "language_loss": 0.78473139,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.8068974,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.692394971847534
    },
    {
      "auxiliary_loss_clip": 0.01182075,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.05639553,
      "balance_loss_mlp": 1.02191997,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.900475940183696,
      "language_loss": 0.70933604,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73146188,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.690005302429199
    },
    {
      "auxiliary_loss_clip": 0.01179605,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.05710316,
      "balance_loss_mlp": 1.02034092,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 1.78805140459522,
      "language_loss": 0.86479604,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.88689053,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.7405459880828857
    },
    {
      "auxiliary_loss_clip": 0.01114697,
      "auxiliary_loss_mlp": 0.01006216,
      "balance_loss_clip": 1.03910267,
      "balance_loss_mlp": 1.00426126,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9375868614761743,
      "language_loss": 0.62494904,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64615822,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.2270288467407227
    },
    {
      "auxiliary_loss_clip": 0.012053,
      "auxiliary_loss_mlp": 0.0103679,
      "balance_loss_clip": 1.06152511,
      "balance_loss_mlp": 1.02549469,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.8975892002793366,
      "language_loss": 0.71647036,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.73889124,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.70109224319458
    },
    {
      "auxiliary_loss_clip": 0.0120023,
      "auxiliary_loss_mlp": 0.0103443,
      "balance_loss_clip": 1.06331241,
      "balance_loss_mlp": 1.02484536,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 3.4584976205344304,
      "language_loss": 0.83939022,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86173683,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.653759479522705
    },
    {
      "auxiliary_loss_clip": 0.01183813,
      "auxiliary_loss_mlp": 0.01032379,
      "balance_loss_clip": 1.0571413,
      "balance_loss_mlp": 1.02229953,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.914152887553662,
      "language_loss": 0.91007984,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93224174,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.683875322341919
    },
    {
      "auxiliary_loss_clip": 0.01194597,
      "auxiliary_loss_mlp": 0.01041035,
      "balance_loss_clip": 1.06340361,
      "balance_loss_mlp": 1.0308907,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.9634868285034153,
      "language_loss": 0.85203809,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87439442,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.6954586505889893
    },
    {
      "auxiliary_loss_clip": 0.01210601,
      "auxiliary_loss_mlp": 0.00902953,
      "balance_loss_clip": 1.06296968,
      "balance_loss_mlp": 1.00208521,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 1.9697951626027008,
      "language_loss": 0.91089022,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.93202567,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.683725118637085
    },
    {
      "auxiliary_loss_clip": 0.01196346,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.0601356,
      "balance_loss_mlp": 1.0228529,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.7544189663420044,
      "language_loss": 0.78530717,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80759227,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.6887428760528564
    },
    {
      "auxiliary_loss_clip": 0.01213669,
      "auxiliary_loss_mlp": 0.01034915,
      "balance_loss_clip": 1.06311107,
      "balance_loss_mlp": 1.02564085,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.8182570393156867,
      "language_loss": 0.91652507,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93901098,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.645364284515381
    },
    {
      "auxiliary_loss_clip": 0.01193344,
      "auxiliary_loss_mlp": 0.01034611,
      "balance_loss_clip": 1.05926061,
      "balance_loss_mlp": 1.02478838,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 2.0552163604319054,
      "language_loss": 0.73223859,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.75451815,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.660205841064453
    },
    {
      "auxiliary_loss_clip": 0.01199387,
      "auxiliary_loss_mlp": 0.01030764,
      "balance_loss_clip": 1.05847883,
      "balance_loss_mlp": 1.02057719,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.051896774396818,
      "language_loss": 0.78851575,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81081724,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.6657862663269043
    },
    {
      "auxiliary_loss_clip": 0.01167122,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.05251658,
      "balance_loss_mlp": 1.02381682,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 1.920172103455171,
      "language_loss": 0.78313017,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80513793,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.8032023906707764
    },
    {
      "auxiliary_loss_clip": 0.01197398,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.06066966,
      "balance_loss_mlp": 1.02024007,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 2.0296913418769913,
      "language_loss": 0.85363257,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87590313,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.6896891593933105
    },
    {
      "auxiliary_loss_clip": 0.01191077,
      "auxiliary_loss_mlp": 0.01033161,
      "balance_loss_clip": 1.05643272,
      "balance_loss_mlp": 1.02330256,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 2.048564534499399,
      "language_loss": 0.83032501,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85256732,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.734123945236206
    },
    {
      "auxiliary_loss_clip": 0.01211767,
      "auxiliary_loss_mlp": 0.0090357,
      "balance_loss_clip": 1.06491947,
      "balance_loss_mlp": 1.00201249,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.9405417210585687,
      "language_loss": 0.74041283,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76156616,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.5873072147369385
    },
    {
      "auxiliary_loss_clip": 0.01177619,
      "auxiliary_loss_mlp": 0.01032484,
      "balance_loss_clip": 1.05460167,
      "balance_loss_mlp": 1.02244091,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.819067080919224,
      "language_loss": 0.68233293,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70443404,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 3.6872966289520264
    },
    {
      "auxiliary_loss_clip": 0.01215555,
      "auxiliary_loss_mlp": 0.01031453,
      "balance_loss_clip": 1.06070817,
      "balance_loss_mlp": 1.02096248,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.9415891388067097,
      "language_loss": 0.69461441,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71708453,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 3.602433919906616
    },
    {
      "auxiliary_loss_clip": 0.01091874,
      "auxiliary_loss_mlp": 0.01001039,
      "balance_loss_clip": 1.02869558,
      "balance_loss_mlp": 0.9989056,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8763640361799088,
      "language_loss": 0.55509627,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57602543,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.1940808296203613
    },
    {
      "auxiliary_loss_clip": 0.01187258,
      "auxiliary_loss_mlp": 0.01027871,
      "balance_loss_clip": 1.05572677,
      "balance_loss_mlp": 1.01860809,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 1.9026969060210608,
      "language_loss": 0.78943056,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.81158185,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 3.5732109546661377
    },
    {
      "auxiliary_loss_clip": 0.01172613,
      "auxiliary_loss_mlp": 0.01031101,
      "balance_loss_clip": 1.0550878,
      "balance_loss_mlp": 1.02204764,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 2.166971487861503,
      "language_loss": 0.82228851,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.84432566,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.7337446212768555
    },
    {
      "auxiliary_loss_clip": 0.01089107,
      "auxiliary_loss_mlp": 0.01002108,
      "balance_loss_clip": 1.02694535,
      "balance_loss_mlp": 1.00028408,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.869443307831769,
      "language_loss": 0.59926784,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.62018001,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.2804787158966064
    },
    {
      "auxiliary_loss_clip": 0.01176762,
      "auxiliary_loss_mlp": 0.01033909,
      "balance_loss_clip": 1.05585217,
      "balance_loss_mlp": 1.02427673,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 3.1873250606528125,
      "language_loss": 0.70246965,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72457635,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.8058512210845947
    },
    {
      "auxiliary_loss_clip": 0.01199315,
      "auxiliary_loss_mlp": 0.01039494,
      "balance_loss_clip": 1.06341875,
      "balance_loss_mlp": 1.02918792,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.2879058725207395,
      "language_loss": 0.83798718,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.86037529,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 3.6426353454589844
    },
    {
      "auxiliary_loss_clip": 0.01186771,
      "auxiliary_loss_mlp": 0.01029145,
      "balance_loss_clip": 1.05847943,
      "balance_loss_mlp": 1.0201509,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.6410398702948403,
      "language_loss": 0.80111629,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82327545,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.7116377353668213
    },
    {
      "auxiliary_loss_clip": 0.01178232,
      "auxiliary_loss_mlp": 0.0102901,
      "balance_loss_clip": 1.05998611,
      "balance_loss_mlp": 1.01981354,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.1491871569316983,
      "language_loss": 0.79015332,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.81222582,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.6883678436279297
    },
    {
      "auxiliary_loss_clip": 0.01198103,
      "auxiliary_loss_mlp": 0.01028822,
      "balance_loss_clip": 1.05764222,
      "balance_loss_mlp": 1.01801515,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.0876984204559363,
      "language_loss": 0.67280316,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69507235,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.6471221446990967
    },
    {
      "auxiliary_loss_clip": 0.01185195,
      "auxiliary_loss_mlp": 0.01028312,
      "balance_loss_clip": 1.05687904,
      "balance_loss_mlp": 1.01883483,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 1.7152524872827655,
      "language_loss": 0.6916638,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71379888,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.6792378425598145
    },
    {
      "auxiliary_loss_clip": 0.01201836,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.06103539,
      "balance_loss_mlp": 1.01954961,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.3191739849705155,
      "language_loss": 0.77550161,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79781139,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.627993106842041
    },
    {
      "auxiliary_loss_clip": 0.01176426,
      "auxiliary_loss_mlp": 0.01034934,
      "balance_loss_clip": 1.05170608,
      "balance_loss_mlp": 1.02557552,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 2.9558907775519567,
      "language_loss": 0.65805584,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.68016946,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.6587955951690674
    },
    {
      "auxiliary_loss_clip": 0.01162873,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.0510447,
      "balance_loss_mlp": 1.01994586,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 1.7374477293736805,
      "language_loss": 0.77874559,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80066788,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.7383298873901367
    },
    {
      "auxiliary_loss_clip": 0.0119485,
      "auxiliary_loss_mlp": 0.01036493,
      "balance_loss_clip": 1.06204188,
      "balance_loss_mlp": 1.02740312,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.5306949578282656,
      "language_loss": 0.73305762,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75537103,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.6792714595794678
    },
    {
      "auxiliary_loss_clip": 0.01207174,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.06036365,
      "balance_loss_mlp": 1.02055979,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.290294151113629,
      "language_loss": 0.75586689,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77823937,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.591432571411133
    },
    {
      "auxiliary_loss_clip": 0.01173927,
      "auxiliary_loss_mlp": 0.01032373,
      "balance_loss_clip": 1.05669308,
      "balance_loss_mlp": 1.02299154,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 2.7961173417602,
      "language_loss": 0.7751261,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79718912,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.8078603744506836
    },
    {
      "auxiliary_loss_clip": 0.01185842,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.05374789,
      "balance_loss_mlp": 1.02434039,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 2.207742568706501,
      "language_loss": 0.8103224,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.83251053,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.6336288452148438
    },
    {
      "auxiliary_loss_clip": 0.0117962,
      "auxiliary_loss_mlp": 0.01030427,
      "balance_loss_clip": 1.05243087,
      "balance_loss_mlp": 1.02118278,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.3201840461361343,
      "language_loss": 0.8029623,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.82506275,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.746082305908203
    },
    {
      "auxiliary_loss_clip": 0.01190083,
      "auxiliary_loss_mlp": 0.01028895,
      "balance_loss_clip": 1.05591321,
      "balance_loss_mlp": 1.01928043,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.082234016461116,
      "language_loss": 0.80447388,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.82666373,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.639875650405884
    },
    {
      "auxiliary_loss_clip": 0.01194752,
      "auxiliary_loss_mlp": 0.01032111,
      "balance_loss_clip": 1.0565598,
      "balance_loss_mlp": 1.0228188,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.813083383859083,
      "language_loss": 0.81102705,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.8332957,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.8470349311828613
    },
    {
      "auxiliary_loss_clip": 0.01189161,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.0590446,
      "balance_loss_mlp": 1.0221113,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 2.1506560054918196,
      "language_loss": 0.80319583,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82539743,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.689082622528076
    },
    {
      "auxiliary_loss_clip": 0.01149581,
      "auxiliary_loss_mlp": 0.01028445,
      "balance_loss_clip": 1.0468235,
      "balance_loss_mlp": 1.0191462,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 2.7073347748970833,
      "language_loss": 0.7104969,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73227715,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.7534687519073486
    },
    {
      "auxiliary_loss_clip": 0.01198336,
      "auxiliary_loss_mlp": 0.00902895,
      "balance_loss_clip": 1.06180453,
      "balance_loss_mlp": 1.00161314,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 1.9553711004809549,
      "language_loss": 0.80239832,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82341063,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.7632014751434326
    },
    {
      "auxiliary_loss_clip": 0.01110085,
      "auxiliary_loss_mlp": 0.01004512,
      "balance_loss_clip": 1.03401327,
      "balance_loss_mlp": 1.00281906,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0061645086657196,
      "language_loss": 0.60352182,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62466776,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.1784629821777344
    },
    {
      "auxiliary_loss_clip": 0.01199206,
      "auxiliary_loss_mlp": 0.01032813,
      "balance_loss_clip": 1.05967557,
      "balance_loss_mlp": 1.02299619,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 2.0920384524400664,
      "language_loss": 0.78370678,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80602694,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 3.6458635330200195
    },
    {
      "auxiliary_loss_clip": 0.01182962,
      "auxiliary_loss_mlp": 0.01027987,
      "balance_loss_clip": 1.0513854,
      "balance_loss_mlp": 1.01879025,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.279253041130976,
      "language_loss": 0.88535404,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90746355,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.706716775894165
    },
    {
      "auxiliary_loss_clip": 0.01180624,
      "auxiliary_loss_mlp": 0.01034363,
      "balance_loss_clip": 1.05909562,
      "balance_loss_mlp": 1.0250591,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.1915892263365735,
      "language_loss": 0.62716448,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.6493144,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 3.6312522888183594
    },
    {
      "auxiliary_loss_clip": 0.01205694,
      "auxiliary_loss_mlp": 0.01039191,
      "balance_loss_clip": 1.06111264,
      "balance_loss_mlp": 1.03049469,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 11.892580413592917,
      "language_loss": 0.85040951,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87285835,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 3.495046854019165
    },
    {
      "auxiliary_loss_clip": 0.01196766,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.05622029,
      "balance_loss_mlp": 1.02461815,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.4351691696858797,
      "language_loss": 0.885548,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90785551,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.7447714805603027
    },
    {
      "auxiliary_loss_clip": 0.01198249,
      "auxiliary_loss_mlp": 0.01033702,
      "balance_loss_clip": 1.06108022,
      "balance_loss_mlp": 1.02390635,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.7445735398959759,
      "language_loss": 0.81820786,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.84052742,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.6411426067352295
    },
    {
      "auxiliary_loss_clip": 0.01164679,
      "auxiliary_loss_mlp": 0.01030902,
      "balance_loss_clip": 1.05228829,
      "balance_loss_mlp": 1.02225971,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 2.7013794619197404,
      "language_loss": 0.76000011,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78195584,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.8288540840148926
    },
    {
      "auxiliary_loss_clip": 0.01196411,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.0585736,
      "balance_loss_mlp": 1.02238035,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.053252575356353,
      "language_loss": 0.86037278,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88265681,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.6346681118011475
    },
    {
      "auxiliary_loss_clip": 0.01168568,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 1.04973841,
      "balance_loss_mlp": 1.02248609,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.843459785693826,
      "language_loss": 0.83993185,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86192894,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 3.6155283451080322
    },
    {
      "auxiliary_loss_clip": 0.01094869,
      "auxiliary_loss_mlp": 0.01004441,
      "balance_loss_clip": 1.03068423,
      "balance_loss_mlp": 1.00253391,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.9025274393532406,
      "language_loss": 0.67931318,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.7003063,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.3143701553344727
    },
    {
      "auxiliary_loss_clip": 0.01186618,
      "auxiliary_loss_mlp": 0.01035998,
      "balance_loss_clip": 1.05893242,
      "balance_loss_mlp": 1.0261569,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 1.9591758315453338,
      "language_loss": 0.86346436,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88569057,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 2.7021501064300537
    },
    {
      "auxiliary_loss_clip": 0.01176245,
      "auxiliary_loss_mlp": 0.01038801,
      "balance_loss_clip": 1.05568945,
      "balance_loss_mlp": 1.02898455,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.973981066764209,
      "language_loss": 0.84914273,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.87129319,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.700782299041748
    },
    {
      "auxiliary_loss_clip": 0.01196774,
      "auxiliary_loss_mlp": 0.01035094,
      "balance_loss_clip": 1.06117713,
      "balance_loss_mlp": 1.02600431,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.7785705748966094,
      "language_loss": 0.66543674,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68775535,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.6747395992279053
    },
    {
      "auxiliary_loss_clip": 0.01196837,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.05905652,
      "balance_loss_mlp": 1.0212152,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 5.729437475520219,
      "language_loss": 0.82145035,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84372991,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.628772497177124
    },
    {
      "auxiliary_loss_clip": 0.0120503,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.05803537,
      "balance_loss_mlp": 1.02173841,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.3351538549123974,
      "language_loss": 0.68495858,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.70732504,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.84049391746521
    },
    {
      "auxiliary_loss_clip": 0.01186131,
      "auxiliary_loss_mlp": 0.01041049,
      "balance_loss_clip": 1.05505252,
      "balance_loss_mlp": 1.03039193,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 1.8701718398661993,
      "language_loss": 0.79729259,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81956446,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.739121198654175
    },
    {
      "auxiliary_loss_clip": 0.01179969,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.05467558,
      "balance_loss_mlp": 1.0170269,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 1.8949575404739667,
      "language_loss": 0.68212861,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70419216,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.650602102279663
    },
    {
      "auxiliary_loss_clip": 0.01186153,
      "auxiliary_loss_mlp": 0.01032385,
      "balance_loss_clip": 1.05584884,
      "balance_loss_mlp": 1.02302098,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 2.155082924414394,
      "language_loss": 0.73086876,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75305414,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.745872735977173
    },
    {
      "auxiliary_loss_clip": 0.01186365,
      "auxiliary_loss_mlp": 0.01030026,
      "balance_loss_clip": 1.05257523,
      "balance_loss_mlp": 1.02009034,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 2.1828511098170402,
      "language_loss": 0.85389042,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87605435,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.6254005432128906
    },
    {
      "auxiliary_loss_clip": 0.01107145,
      "auxiliary_loss_mlp": 0.01012647,
      "balance_loss_clip": 1.05446339,
      "balance_loss_mlp": 1.010692,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7337322760497267,
      "language_loss": 0.55928481,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58048272,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.190609931945801
    },
    {
      "auxiliary_loss_clip": 0.01151971,
      "auxiliary_loss_mlp": 0.01032558,
      "balance_loss_clip": 1.04748845,
      "balance_loss_mlp": 1.02230012,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.0935893466135993,
      "language_loss": 0.77635264,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.79819793,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.7146265506744385
    },
    {
      "auxiliary_loss_clip": 0.01187626,
      "auxiliary_loss_mlp": 0.0103004,
      "balance_loss_clip": 1.05650711,
      "balance_loss_mlp": 1.0205214,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 1.9780439347969394,
      "language_loss": 0.80096459,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.82314122,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.720468759536743
    },
    {
      "auxiliary_loss_clip": 0.01199656,
      "auxiliary_loss_mlp": 0.00903155,
      "balance_loss_clip": 1.05874515,
      "balance_loss_mlp": 1.00169873,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 2.1744146507934357,
      "language_loss": 0.78893965,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.80996776,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.641829490661621
    },
    {
      "auxiliary_loss_clip": 0.0117875,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.05552411,
      "balance_loss_mlp": 1.02406144,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.639041756123092,
      "language_loss": 0.84988546,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87201208,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.709374189376831
    },
    {
      "auxiliary_loss_clip": 0.01188154,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.0556047,
      "balance_loss_mlp": 1.0254705,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.5835775627878115,
      "language_loss": 0.65183389,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67406416,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.7548141479492188
    },
    {
      "auxiliary_loss_clip": 0.01164317,
      "auxiliary_loss_mlp": 0.01024824,
      "balance_loss_clip": 1.05255187,
      "balance_loss_mlp": 1.01610422,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.5401084961799247,
      "language_loss": 0.82502019,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84691155,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.76285719871521
    },
    {
      "auxiliary_loss_clip": 0.01190928,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.06060696,
      "balance_loss_mlp": 1.02472818,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 2.6783315464901025,
      "language_loss": 0.78990567,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.81215632,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 3.7371702194213867
    },
    {
      "auxiliary_loss_clip": 0.01171945,
      "auxiliary_loss_mlp": 0.0103369,
      "balance_loss_clip": 1.05518842,
      "balance_loss_mlp": 1.02400994,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 1.8896521047436878,
      "language_loss": 0.67971587,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70177233,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.6475467681884766
    },
    {
      "auxiliary_loss_clip": 0.01186294,
      "auxiliary_loss_mlp": 0.01033039,
      "balance_loss_clip": 1.05680764,
      "balance_loss_mlp": 1.02375817,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 1.8306148089109746,
      "language_loss": 0.73321044,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75540376,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 3.6018450260162354
    },
    {
      "auxiliary_loss_clip": 0.0119872,
      "auxiliary_loss_mlp": 0.01038314,
      "balance_loss_clip": 1.06286907,
      "balance_loss_mlp": 1.02840745,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 2.145223697072955,
      "language_loss": 0.75362819,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77599853,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 3.6518497467041016
    },
    {
      "auxiliary_loss_clip": 0.01190823,
      "auxiliary_loss_mlp": 0.01032426,
      "balance_loss_clip": 1.05740869,
      "balance_loss_mlp": 1.0232228,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 1.7984263609220872,
      "language_loss": 0.70864427,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.7308768,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.753204584121704
    },
    {
      "auxiliary_loss_clip": 0.01200127,
      "auxiliary_loss_mlp": 0.00903458,
      "balance_loss_clip": 1.05638838,
      "balance_loss_mlp": 1.00168991,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.0085844283829,
      "language_loss": 0.70568186,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72671771,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.5861010551452637
    },
    {
      "auxiliary_loss_clip": 0.01154027,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.05640197,
      "balance_loss_mlp": 1.02180624,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 1.8759976905325502,
      "language_loss": 0.73093057,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.7527914,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.972478151321411
    },
    {
      "auxiliary_loss_clip": 0.01188283,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.05865562,
      "balance_loss_mlp": 1.01807833,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.5015231086932714,
      "language_loss": 0.774562,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.79672599,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.6967577934265137
    },
    {
      "auxiliary_loss_clip": 0.0120975,
      "auxiliary_loss_mlp": 0.00903995,
      "balance_loss_clip": 1.06116688,
      "balance_loss_mlp": 1.00175571,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.483951017051353,
      "language_loss": 0.73445833,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.7555958,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.6364076137542725
    },
    {
      "auxiliary_loss_clip": 0.01201495,
      "auxiliary_loss_mlp": 0.00902632,
      "balance_loss_clip": 1.06157064,
      "balance_loss_mlp": 1.00158811,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 2.0812940199463603,
      "language_loss": 0.81483221,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83587348,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 3.606489896774292
    },
    {
      "auxiliary_loss_clip": 0.01186413,
      "auxiliary_loss_mlp": 0.01029895,
      "balance_loss_clip": 1.05726147,
      "balance_loss_mlp": 1.02084136,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 2.5819191807223336,
      "language_loss": 0.89572775,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91789079,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.669299840927124
    },
    {
      "auxiliary_loss_clip": 0.01210429,
      "auxiliary_loss_mlp": 0.01034493,
      "balance_loss_clip": 1.06388414,
      "balance_loss_mlp": 1.02430701,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 11.888334082817732,
      "language_loss": 0.67188841,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69433761,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.726454496383667
    },
    {
      "auxiliary_loss_clip": 0.01174726,
      "auxiliary_loss_mlp": 0.01035157,
      "balance_loss_clip": 1.05628228,
      "balance_loss_mlp": 1.0253104,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.0580025431698274,
      "language_loss": 0.88413697,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90623581,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.6985647678375244
    },
    {
      "auxiliary_loss_clip": 0.01160224,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.04821777,
      "balance_loss_mlp": 1.01882207,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.894149982949911,
      "language_loss": 0.64399385,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66588491,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.882050037384033
    },
    {
      "auxiliary_loss_clip": 0.01181657,
      "auxiliary_loss_mlp": 0.01034727,
      "balance_loss_clip": 1.05788112,
      "balance_loss_mlp": 1.02586937,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 13.116866196634179,
      "language_loss": 0.64526248,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66742635,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.758819818496704
    },
    {
      "auxiliary_loss_clip": 0.01166224,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05269873,
      "balance_loss_mlp": 1.01664186,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 2.1766491209974306,
      "language_loss": 0.68953538,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.71146172,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.912200689315796
    },
    {
      "auxiliary_loss_clip": 0.01171628,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.05606878,
      "balance_loss_mlp": 1.01612854,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.2754240229260887,
      "language_loss": 0.85608464,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87805247,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.7032511234283447
    },
    {
      "auxiliary_loss_clip": 0.01211065,
      "auxiliary_loss_mlp": 0.01035014,
      "balance_loss_clip": 1.06078506,
      "balance_loss_mlp": 1.02421927,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 2.0228299576497957,
      "language_loss": 0.72120714,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.7436679,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.681993007659912
    },
    {
      "auxiliary_loss_clip": 0.01198723,
      "auxiliary_loss_mlp": 0.01031454,
      "balance_loss_clip": 1.05928087,
      "balance_loss_mlp": 1.02161336,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.803827590910505,
      "language_loss": 0.77352822,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.79583001,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.648585081100464
    },
    {
      "auxiliary_loss_clip": 0.01131063,
      "auxiliary_loss_mlp": 0.00904095,
      "balance_loss_clip": 1.04965675,
      "balance_loss_mlp": 1.00188899,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 2.0073647834809316,
      "language_loss": 0.80291831,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.82326996,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.8583548069000244
    },
    {
      "auxiliary_loss_clip": 0.0111103,
      "auxiliary_loss_mlp": 0.01006228,
      "balance_loss_clip": 1.0387013,
      "balance_loss_mlp": 1.00463033,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8612602101148975,
      "language_loss": 0.61214352,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63331604,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.2868950366973877
    },
    {
      "auxiliary_loss_clip": 0.01199043,
      "auxiliary_loss_mlp": 0.01031082,
      "balance_loss_clip": 1.06079733,
      "balance_loss_mlp": 1.02227235,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 11.658685833439717,
      "language_loss": 0.76968193,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.79198319,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.7108161449432373
    },
    {
      "auxiliary_loss_clip": 0.01181638,
      "auxiliary_loss_mlp": 0.01033514,
      "balance_loss_clip": 1.05580401,
      "balance_loss_mlp": 1.02394783,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.170563125411067,
      "language_loss": 0.78436273,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80651426,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.8531293869018555
    },
    {
      "auxiliary_loss_clip": 0.0120444,
      "auxiliary_loss_mlp": 0.01038259,
      "balance_loss_clip": 1.06053746,
      "balance_loss_mlp": 1.02864504,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 2.576126349603904,
      "language_loss": 0.87987566,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90230268,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.6501758098602295
    },
    {
      "auxiliary_loss_clip": 0.01196481,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.06198907,
      "balance_loss_mlp": 1.02552104,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.04632087250135,
      "language_loss": 0.87967086,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90198439,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.6853108406066895
    },
    {
      "auxiliary_loss_clip": 0.01196816,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.06008363,
      "balance_loss_mlp": 1.02280688,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 2.426788459451446,
      "language_loss": 0.77988195,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.80217105,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.665337085723877
    },
    {
      "auxiliary_loss_clip": 0.01104842,
      "auxiliary_loss_mlp": 0.01002041,
      "balance_loss_clip": 1.03435588,
      "balance_loss_mlp": 1.00050354,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7724853986770005,
      "language_loss": 0.59704882,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61811769,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 3.0876283645629883
    },
    {
      "auxiliary_loss_clip": 0.01164957,
      "auxiliary_loss_mlp": 0.01033968,
      "balance_loss_clip": 1.0561198,
      "balance_loss_mlp": 1.02462757,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.7582554016868306,
      "language_loss": 0.76850021,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.79048944,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 3.732530355453491
    },
    {
      "auxiliary_loss_clip": 0.01192163,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 1.05635583,
      "balance_loss_mlp": 1.01897848,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 4.237982895431636,
      "language_loss": 0.80955607,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83175713,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.6889700889587402
    },
    {
      "auxiliary_loss_clip": 0.01189301,
      "auxiliary_loss_mlp": 0.01032642,
      "balance_loss_clip": 1.05619431,
      "balance_loss_mlp": 1.02373695,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 2.142074608925352,
      "language_loss": 0.77241933,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79463875,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 3.599355697631836
    },
    {
      "auxiliary_loss_clip": 0.01188348,
      "auxiliary_loss_mlp": 0.01037907,
      "balance_loss_clip": 1.05648446,
      "balance_loss_mlp": 1.02893698,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.4798258095965553,
      "language_loss": 0.77856171,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.80082428,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 3.7078986167907715
    },
    {
      "auxiliary_loss_clip": 0.01138449,
      "auxiliary_loss_mlp": 0.01033501,
      "balance_loss_clip": 1.05103862,
      "balance_loss_mlp": 1.02413154,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 2.635723180661539,
      "language_loss": 0.77784169,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.79956126,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.749753952026367
    },
    {
      "auxiliary_loss_clip": 0.0120387,
      "auxiliary_loss_mlp": 0.01033717,
      "balance_loss_clip": 1.06034982,
      "balance_loss_mlp": 1.02515781,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 2.0592717084567007,
      "language_loss": 0.79451168,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81688756,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.666654109954834
    },
    {
      "auxiliary_loss_clip": 0.01185686,
      "auxiliary_loss_mlp": 0.01032579,
      "balance_loss_clip": 1.05902338,
      "balance_loss_mlp": 1.02297032,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 2.4478305756468397,
      "language_loss": 0.86718506,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88936764,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.636256456375122
    },
    {
      "auxiliary_loss_clip": 0.01166946,
      "auxiliary_loss_mlp": 0.01035614,
      "balance_loss_clip": 1.05765724,
      "balance_loss_mlp": 1.02616072,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.859945429831035,
      "language_loss": 0.78230894,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80433452,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.8346874713897705
    },
    {
      "auxiliary_loss_clip": 0.01084926,
      "auxiliary_loss_mlp": 0.00892559,
      "balance_loss_clip": 1.02464366,
      "balance_loss_mlp": 1.00044274,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.697274957155989,
      "language_loss": 0.54336047,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56313527,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 4.186205863952637
    },
    {
      "auxiliary_loss_clip": 0.01174797,
      "auxiliary_loss_mlp": 0.01032022,
      "balance_loss_clip": 1.0529691,
      "balance_loss_mlp": 1.0228188,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.8226333788686575,
      "language_loss": 0.89223504,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91430324,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.6636734008789062
    },
    {
      "auxiliary_loss_clip": 0.01188401,
      "auxiliary_loss_mlp": 0.01034976,
      "balance_loss_clip": 1.05709827,
      "balance_loss_mlp": 1.02581453,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 2.402968971633432,
      "language_loss": 0.75570214,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.77793598,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.7145440578460693
    },
    {
      "auxiliary_loss_clip": 0.01161049,
      "auxiliary_loss_mlp": 0.01042274,
      "balance_loss_clip": 1.05343199,
      "balance_loss_mlp": 1.03264809,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 2.1246980565908986,
      "language_loss": 0.66014075,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68217397,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.8339715003967285
    },
    {
      "auxiliary_loss_clip": 0.01110744,
      "auxiliary_loss_mlp": 0.01004229,
      "balance_loss_clip": 1.03188705,
      "balance_loss_mlp": 1.00257194,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.9748559863714734,
      "language_loss": 0.6066432,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62779295,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.0387275218963623
    },
    {
      "auxiliary_loss_clip": 0.01190835,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.05629873,
      "balance_loss_mlp": 1.02446342,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.3914086988954657,
      "language_loss": 0.92321026,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.9454577,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.721231698989868
    },
    {
      "auxiliary_loss_clip": 0.01201789,
      "auxiliary_loss_mlp": 0.01033171,
      "balance_loss_clip": 1.05881596,
      "balance_loss_mlp": 1.02375984,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.5217548441338207,
      "language_loss": 0.81634694,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.8386966,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.6044185161590576
    },
    {
      "auxiliary_loss_clip": 0.01189175,
      "auxiliary_loss_mlp": 0.01029406,
      "balance_loss_clip": 1.06034958,
      "balance_loss_mlp": 1.02009583,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.832595285470781,
      "language_loss": 0.86944401,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89162976,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.7282636165618896
    },
    {
      "auxiliary_loss_clip": 0.01206603,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.05965185,
      "balance_loss_mlp": 1.01783824,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.154129415208991,
      "language_loss": 0.82782936,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.8501637,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.5872018337249756
    },
    {
      "auxiliary_loss_clip": 0.01178436,
      "auxiliary_loss_mlp": 0.0103225,
      "balance_loss_clip": 1.05381787,
      "balance_loss_mlp": 1.02285647,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 2.8305667923735633,
      "language_loss": 0.81102377,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.8331306,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.6777849197387695
    },
    {
      "auxiliary_loss_clip": 0.010938,
      "auxiliary_loss_mlp": 0.01004846,
      "balance_loss_clip": 1.02513099,
      "balance_loss_mlp": 1.00340319,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 1.192800038053706,
      "language_loss": 0.51801598,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53900242,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.313661575317383
    },
    {
      "auxiliary_loss_clip": 0.0117172,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.05246902,
      "balance_loss_mlp": 1.02305949,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.6071165681394035,
      "language_loss": 0.79859877,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.82063431,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.7141454219818115
    },
    {
      "auxiliary_loss_clip": 0.01093998,
      "auxiliary_loss_mlp": 0.01001037,
      "balance_loss_clip": 1.02649271,
      "balance_loss_mlp": 0.99964821,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8132929927100447,
      "language_loss": 0.62488157,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64583188,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.298677682876587
    },
    {
      "auxiliary_loss_clip": 0.01195209,
      "auxiliary_loss_mlp": 0.01029039,
      "balance_loss_clip": 1.05737495,
      "balance_loss_mlp": 1.01973486,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.6408717909339807,
      "language_loss": 0.63717484,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.65941733,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.646395206451416
    },
    {
      "auxiliary_loss_clip": 0.0118547,
      "auxiliary_loss_mlp": 0.0103442,
      "balance_loss_clip": 1.05634069,
      "balance_loss_mlp": 1.02540159,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.2598403566299474,
      "language_loss": 0.76086539,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78306431,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.8021700382232666
    },
    {
      "auxiliary_loss_clip": 0.01195599,
      "auxiliary_loss_mlp": 0.01032201,
      "balance_loss_clip": 1.05938053,
      "balance_loss_mlp": 1.02361822,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.46010188050185,
      "language_loss": 0.82291269,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84519064,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.658290147781372
    },
    {
      "auxiliary_loss_clip": 0.01085547,
      "auxiliary_loss_mlp": 0.00892095,
      "balance_loss_clip": 1.02368152,
      "balance_loss_mlp": 1.00020742,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8657898097905679,
      "language_loss": 0.63711476,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65689123,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.2784740924835205
    },
    {
      "auxiliary_loss_clip": 0.01175448,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.0577141,
      "balance_loss_mlp": 1.02305865,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 2.2284222458583742,
      "language_loss": 0.80850345,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.83058167,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.7692105770111084
    },
    {
      "auxiliary_loss_clip": 0.01093229,
      "auxiliary_loss_mlp": 0.01001854,
      "balance_loss_clip": 1.02650261,
      "balance_loss_mlp": 1.00053048,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.8003528919840678,
      "language_loss": 0.60423434,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62518513,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 4.089259147644043
    },
    {
      "auxiliary_loss_clip": 0.01200695,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.05692601,
      "balance_loss_mlp": 1.02631879,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.7453524814837404,
      "language_loss": 0.7944361,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81679511,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.634343385696411
    },
    {
      "auxiliary_loss_clip": 0.01169075,
      "auxiliary_loss_mlp": 0.01033001,
      "balance_loss_clip": 1.05098736,
      "balance_loss_mlp": 1.02356601,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.362858559553279,
      "language_loss": 0.63205016,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65407085,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.7710118293762207
    },
    {
      "auxiliary_loss_clip": 0.01195416,
      "auxiliary_loss_mlp": 0.01032524,
      "balance_loss_clip": 1.05890179,
      "balance_loss_mlp": 1.02374423,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.8815495431895086,
      "language_loss": 0.77531183,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.79759121,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 4.500869035720825
    },
    {
      "auxiliary_loss_clip": 0.01168811,
      "auxiliary_loss_mlp": 0.01041623,
      "balance_loss_clip": 1.05307066,
      "balance_loss_mlp": 1.03224683,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.6801845218491187,
      "language_loss": 0.78444242,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.80654669,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.6846675872802734
    },
    {
      "auxiliary_loss_clip": 0.0116294,
      "auxiliary_loss_mlp": 0.01034258,
      "balance_loss_clip": 1.05359387,
      "balance_loss_mlp": 1.02546644,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.7410820244795555,
      "language_loss": 0.78037953,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80235147,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.733917713165283
    },
    {
      "auxiliary_loss_clip": 0.01204255,
      "auxiliary_loss_mlp": 0.01032716,
      "balance_loss_clip": 1.05706596,
      "balance_loss_mlp": 1.02385259,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 1.7200068826099286,
      "language_loss": 0.82945037,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85182005,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.675837993621826
    },
    {
      "auxiliary_loss_clip": 0.01167789,
      "auxiliary_loss_mlp": 0.00903451,
      "balance_loss_clip": 1.05384421,
      "balance_loss_mlp": 1.00194192,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 2.0048760979744507,
      "language_loss": 0.75351524,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77422762,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.7316200733184814
    },
    {
      "auxiliary_loss_clip": 0.01200179,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.06174159,
      "balance_loss_mlp": 1.02512312,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 1.879821260605745,
      "language_loss": 0.84523201,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86758626,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 3.6337263584136963
    },
    {
      "auxiliary_loss_clip": 0.0119452,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.05822384,
      "balance_loss_mlp": 1.02011502,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 3.171715389943028,
      "language_loss": 0.66642106,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68865842,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.601653814315796
    },
    {
      "auxiliary_loss_clip": 0.01184675,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05429101,
      "balance_loss_mlp": 1.02066803,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.8922133691859044,
      "language_loss": 0.80391449,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.8260591,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.6082088947296143
    },
    {
      "auxiliary_loss_clip": 0.01104985,
      "auxiliary_loss_mlp": 0.01004182,
      "balance_loss_clip": 1.02754676,
      "balance_loss_mlp": 1.00275159,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7789025242126907,
      "language_loss": 0.55911481,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58020651,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.145989179611206
    },
    {
      "auxiliary_loss_clip": 0.01187327,
      "auxiliary_loss_mlp": 0.01025151,
      "balance_loss_clip": 1.05519617,
      "balance_loss_mlp": 1.01627004,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 2.2231753373400243,
      "language_loss": 0.80604172,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.82816648,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.6569502353668213
    },
    {
      "auxiliary_loss_clip": 0.01208972,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.05996537,
      "balance_loss_mlp": 1.01979089,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 2.4560100494437296,
      "language_loss": 0.78883868,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.81122816,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.6212356090545654
    },
    {
      "auxiliary_loss_clip": 0.01197475,
      "auxiliary_loss_mlp": 0.01029925,
      "balance_loss_clip": 1.06479275,
      "balance_loss_mlp": 1.02065682,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 1.8238915115291356,
      "language_loss": 0.87370199,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89597595,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.612595319747925
    },
    {
      "auxiliary_loss_clip": 0.01174187,
      "auxiliary_loss_mlp": 0.01034117,
      "balance_loss_clip": 1.05233085,
      "balance_loss_mlp": 1.02422917,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.1723130311310035,
      "language_loss": 0.90667558,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92875868,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.690880298614502
    },
    {
      "auxiliary_loss_clip": 0.01184219,
      "auxiliary_loss_mlp": 0.01032317,
      "balance_loss_clip": 1.05309439,
      "balance_loss_mlp": 1.02300096,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.014300099620257,
      "language_loss": 0.78465509,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80682051,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.623910427093506
    },
    {
      "auxiliary_loss_clip": 0.01194214,
      "auxiliary_loss_mlp": 0.00903506,
      "balance_loss_clip": 1.05799246,
      "balance_loss_mlp": 1.00185704,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.187903883464432,
      "language_loss": 0.8011148,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82209194,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.703885555267334
    },
    {
      "auxiliary_loss_clip": 0.01196,
      "auxiliary_loss_mlp": 0.01037821,
      "balance_loss_clip": 1.06031442,
      "balance_loss_mlp": 1.02866554,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 1.8457475170454012,
      "language_loss": 0.72246695,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74480522,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.615886926651001
    },
    {
      "auxiliary_loss_clip": 0.01195374,
      "auxiliary_loss_mlp": 0.00902717,
      "balance_loss_clip": 1.06050253,
      "balance_loss_mlp": 1.00190091,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.5921107825454455,
      "language_loss": 0.72211862,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74309957,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.7157177925109863
    },
    {
      "auxiliary_loss_clip": 0.01160461,
      "auxiliary_loss_mlp": 0.01036823,
      "balance_loss_clip": 1.05283689,
      "balance_loss_mlp": 1.0268451,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.2268183982906016,
      "language_loss": 0.84861934,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.87059224,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 2.7272114753723145
    },
    {
      "auxiliary_loss_clip": 0.01207334,
      "auxiliary_loss_mlp": 0.0090374,
      "balance_loss_clip": 1.06273413,
      "balance_loss_mlp": 1.00195956,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 2.0695229838736977,
      "language_loss": 0.65834939,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67946017,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.631687641143799
    },
    {
      "auxiliary_loss_clip": 0.01192896,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.06172514,
      "balance_loss_mlp": 1.02042782,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.647332477115734,
      "language_loss": 0.82664353,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84886718,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.6586883068084717
    },
    {
      "auxiliary_loss_clip": 0.01187907,
      "auxiliary_loss_mlp": 0.01031776,
      "balance_loss_clip": 1.05868042,
      "balance_loss_mlp": 1.02296066,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.7033542003686795,
      "language_loss": 0.70617956,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.72837639,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.6487607955932617
    },
    {
      "auxiliary_loss_clip": 0.01179008,
      "auxiliary_loss_mlp": 0.01033242,
      "balance_loss_clip": 1.05477524,
      "balance_loss_mlp": 1.02299023,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 3.2896260907304984,
      "language_loss": 0.7418499,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.7639724,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.662156343460083
    },
    {
      "auxiliary_loss_clip": 0.01196544,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.05795801,
      "balance_loss_mlp": 1.02532983,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.9358531007955502,
      "language_loss": 0.87779391,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.90009773,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.651353120803833
    },
    {
      "auxiliary_loss_clip": 0.01209146,
      "auxiliary_loss_mlp": 0.01032558,
      "balance_loss_clip": 1.06170917,
      "balance_loss_mlp": 1.02285385,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.7684467828479442,
      "language_loss": 0.82278955,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.84520656,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.6280291080474854
    },
    {
      "auxiliary_loss_clip": 0.01174069,
      "auxiliary_loss_mlp": 0.01033548,
      "balance_loss_clip": 1.05613112,
      "balance_loss_mlp": 1.02507758,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 3.682246865415019,
      "language_loss": 0.82384002,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84591621,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 3.659541368484497
    },
    {
      "auxiliary_loss_clip": 0.01190675,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.05805957,
      "balance_loss_mlp": 1.02337265,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.7596687744743056,
      "language_loss": 0.75568062,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.77791756,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.800981283187866
    },
    {
      "auxiliary_loss_clip": 0.01197467,
      "auxiliary_loss_mlp": 0.01033338,
      "balance_loss_clip": 1.06028414,
      "balance_loss_mlp": 1.02417672,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.209669559594998,
      "language_loss": 0.77352315,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.7958312,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.721200704574585
    },
    {
      "auxiliary_loss_clip": 0.01188546,
      "auxiliary_loss_mlp": 0.01031474,
      "balance_loss_clip": 1.0565412,
      "balance_loss_mlp": 1.02166355,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.9669298556855161,
      "language_loss": 0.77691746,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79911762,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 4.514526605606079
    },
    {
      "auxiliary_loss_clip": 0.01201376,
      "auxiliary_loss_mlp": 0.01040469,
      "balance_loss_clip": 1.06447864,
      "balance_loss_mlp": 1.03103316,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.2750288328734967,
      "language_loss": 0.76462203,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78704047,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.681734561920166
    },
    {
      "auxiliary_loss_clip": 0.01180368,
      "auxiliary_loss_mlp": 0.01032818,
      "balance_loss_clip": 1.05885696,
      "balance_loss_mlp": 1.02301931,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.8831563850181885,
      "language_loss": 0.80038822,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82252014,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.782731533050537
    },
    {
      "auxiliary_loss_clip": 0.01201889,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.06620014,
      "balance_loss_mlp": 1.01966989,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.545462129012904,
      "language_loss": 0.81425273,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83656108,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.6295840740203857
    },
    {
      "auxiliary_loss_clip": 0.01183128,
      "auxiliary_loss_mlp": 0.00903491,
      "balance_loss_clip": 1.05655813,
      "balance_loss_mlp": 1.00200117,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 2.32932077430609,
      "language_loss": 0.67947328,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.7003395,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.6980397701263428
    },
    {
      "auxiliary_loss_clip": 0.01194701,
      "auxiliary_loss_mlp": 0.01033119,
      "balance_loss_clip": 1.0592922,
      "balance_loss_mlp": 1.02377903,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 4.407228455819127,
      "language_loss": 0.7531724,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77545059,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 3.587672472000122
    },
    {
      "auxiliary_loss_clip": 0.01193991,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.06002474,
      "balance_loss_mlp": 1.01966524,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.4393135584246446,
      "language_loss": 0.791848,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81407487,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.644929885864258
    },
    {
      "auxiliary_loss_clip": 0.01196049,
      "auxiliary_loss_mlp": 0.01031365,
      "balance_loss_clip": 1.05736363,
      "balance_loss_mlp": 1.02172089,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.1612121715967074,
      "language_loss": 0.78630275,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80857688,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.63873028755188
    },
    {
      "auxiliary_loss_clip": 0.01187596,
      "auxiliary_loss_mlp": 0.01035905,
      "balance_loss_clip": 1.06109858,
      "balance_loss_mlp": 1.02671957,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 2.5610860235595387,
      "language_loss": 0.74174231,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76397735,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.7427597045898438
    },
    {
      "auxiliary_loss_clip": 0.01159275,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.05245245,
      "balance_loss_mlp": 1.02026975,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 2.2529823471121926,
      "language_loss": 0.77339494,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.7952857,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.7331790924072266
    },
    {
      "auxiliary_loss_clip": 0.0120801,
      "auxiliary_loss_mlp": 0.01033259,
      "balance_loss_clip": 1.06010473,
      "balance_loss_mlp": 1.02397895,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 1.7772496820623978,
      "language_loss": 0.82384682,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84625947,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.703894853591919
    },
    {
      "auxiliary_loss_clip": 0.01194835,
      "auxiliary_loss_mlp": 0.00904359,
      "balance_loss_clip": 1.06589556,
      "balance_loss_mlp": 1.00200033,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.86949218494292,
      "language_loss": 0.81299537,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83398724,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.721684217453003
    },
    {
      "auxiliary_loss_clip": 0.01184986,
      "auxiliary_loss_mlp": 0.01033217,
      "balance_loss_clip": 1.05849314,
      "balance_loss_mlp": 1.02329874,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 3.4897928769829316,
      "language_loss": 0.77652514,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79870713,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.6775057315826416
    },
    {
      "auxiliary_loss_clip": 0.0118632,
      "auxiliary_loss_mlp": 0.00904061,
      "balance_loss_clip": 1.0630635,
      "balance_loss_mlp": 1.00208855,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.5545243009642453,
      "language_loss": 0.85066342,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87156725,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.6840097904205322
    },
    {
      "auxiliary_loss_clip": 0.01208877,
      "auxiliary_loss_mlp": 0.01034482,
      "balance_loss_clip": 1.06280637,
      "balance_loss_mlp": 1.02502251,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.825113065988909,
      "language_loss": 0.78729165,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80972528,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.5654046535491943
    },
    {
      "auxiliary_loss_clip": 0.01184472,
      "auxiliary_loss_mlp": 0.00904059,
      "balance_loss_clip": 1.05800104,
      "balance_loss_mlp": 1.00190496,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.6575434736946444,
      "language_loss": 0.69284022,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71372557,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.6788318157196045
    },
    {
      "auxiliary_loss_clip": 0.0116856,
      "auxiliary_loss_mlp": 0.0103094,
      "balance_loss_clip": 1.05226386,
      "balance_loss_mlp": 1.02172506,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.0292840243693124,
      "language_loss": 0.61244863,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63444364,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 2.730424404144287
    },
    {
      "auxiliary_loss_clip": 0.01171386,
      "auxiliary_loss_mlp": 0.01035543,
      "balance_loss_clip": 1.05421829,
      "balance_loss_mlp": 1.02588701,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.048488839053899,
      "language_loss": 0.75973397,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.78180331,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.7042236328125
    },
    {
      "auxiliary_loss_clip": 0.01186767,
      "auxiliary_loss_mlp": 0.00902895,
      "balance_loss_clip": 1.0591166,
      "balance_loss_mlp": 1.00194764,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 3.195753646208371,
      "language_loss": 0.80340552,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.82430208,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.694183826446533
    },
    {
      "auxiliary_loss_clip": 0.01093225,
      "auxiliary_loss_mlp": 0.01004227,
      "balance_loss_clip": 1.03048682,
      "balance_loss_mlp": 1.00268888,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8936198568645668,
      "language_loss": 0.70613444,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72710896,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.263310432434082
    },
    {
      "auxiliary_loss_clip": 0.01189618,
      "auxiliary_loss_mlp": 0.01034374,
      "balance_loss_clip": 1.05852842,
      "balance_loss_mlp": 1.02480745,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.7414058590333739,
      "language_loss": 0.78518224,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.8074221,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.703946828842163
    },
    {
      "auxiliary_loss_clip": 0.011925,
      "auxiliary_loss_mlp": 0.01034158,
      "balance_loss_clip": 1.05457997,
      "balance_loss_mlp": 1.02468717,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.290147283951093,
      "language_loss": 0.82234699,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84461355,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.6151723861694336
    },
    {
      "auxiliary_loss_clip": 0.0119869,
      "auxiliary_loss_mlp": 0.01035306,
      "balance_loss_clip": 1.05970025,
      "balance_loss_mlp": 1.02621627,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 2.0676795719352343,
      "language_loss": 0.82944489,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.85178483,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.70466685295105
    },
    {
      "auxiliary_loss_clip": 0.01175245,
      "auxiliary_loss_mlp": 0.01036,
      "balance_loss_clip": 1.05437636,
      "balance_loss_mlp": 1.02630186,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.0817697407255555,
      "language_loss": 0.74874401,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.7708565,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.702921152114868
    },
    {
      "auxiliary_loss_clip": 0.0119341,
      "auxiliary_loss_mlp": 0.01043226,
      "balance_loss_clip": 1.06028676,
      "balance_loss_mlp": 1.03389192,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.3886422841343276,
      "language_loss": 0.7680366,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.79040289,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 3.666926860809326
    },
    {
      "auxiliary_loss_clip": 0.01210331,
      "auxiliary_loss_mlp": 0.01031727,
      "balance_loss_clip": 1.06451893,
      "balance_loss_mlp": 1.02258992,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.029144380658831,
      "language_loss": 0.80902612,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.83144671,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 3.647350549697876
    },
    {
      "auxiliary_loss_clip": 0.01208244,
      "auxiliary_loss_mlp": 0.01037138,
      "balance_loss_clip": 1.06332147,
      "balance_loss_mlp": 1.02778018,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.178880441834486,
      "language_loss": 0.87739569,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.89984953,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 3.4470632076263428
    },
    {
      "auxiliary_loss_clip": 0.01197123,
      "auxiliary_loss_mlp": 0.00904083,
      "balance_loss_clip": 1.05941629,
      "balance_loss_mlp": 1.00209141,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.4040544392711896,
      "language_loss": 0.78530425,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80631626,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.59653639793396
    },
    {
      "auxiliary_loss_clip": 0.01178848,
      "auxiliary_loss_mlp": 0.01034905,
      "balance_loss_clip": 1.05641294,
      "balance_loss_mlp": 1.02566636,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.162695581859936,
      "language_loss": 0.8951304,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.91726792,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.6034748554229736
    },
    {
      "auxiliary_loss_clip": 0.01200552,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.06365848,
      "balance_loss_mlp": 1.02263856,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.5244930663751386,
      "language_loss": 0.8428579,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86518431,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.7438127994537354
    },
    {
      "auxiliary_loss_clip": 0.01199164,
      "auxiliary_loss_mlp": 0.01039423,
      "balance_loss_clip": 1.06588447,
      "balance_loss_mlp": 1.02857471,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 2.0813036943953587,
      "language_loss": 0.81551397,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83789986,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.6949291229248047
    },
    {
      "auxiliary_loss_clip": 0.01188001,
      "auxiliary_loss_mlp": 0.01039232,
      "balance_loss_clip": 1.05781066,
      "balance_loss_mlp": 1.02899241,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.501368465388325,
      "language_loss": 0.77669317,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.79896557,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 3.6216933727264404
    },
    {
      "auxiliary_loss_clip": 0.01200639,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.0633359,
      "balance_loss_mlp": 1.01832771,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 3.037999134346835,
      "language_loss": 0.83374059,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.85602272,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.72629714012146
    },
    {
      "auxiliary_loss_clip": 0.01181915,
      "auxiliary_loss_mlp": 0.01035877,
      "balance_loss_clip": 1.06292367,
      "balance_loss_mlp": 1.02639413,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.572622806264591,
      "language_loss": 0.79036456,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.8125425,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.7886595726013184
    },
    {
      "auxiliary_loss_clip": 0.01178923,
      "auxiliary_loss_mlp": 0.01028422,
      "balance_loss_clip": 1.05935764,
      "balance_loss_mlp": 1.01836693,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 2.048059819406276,
      "language_loss": 0.76697743,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78905094,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.665203332901001
    },
    {
      "auxiliary_loss_clip": 0.01212214,
      "auxiliary_loss_mlp": 0.01036427,
      "balance_loss_clip": 1.06583798,
      "balance_loss_mlp": 1.02712274,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 2.5353338491428534,
      "language_loss": 0.90634644,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92883283,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.639148473739624
    },
    {
      "auxiliary_loss_clip": 0.0119824,
      "auxiliary_loss_mlp": 0.01034103,
      "balance_loss_clip": 1.06377721,
      "balance_loss_mlp": 1.02448273,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.2503686858753107,
      "language_loss": 0.85125959,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87358308,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.629194498062134
    },
    {
      "auxiliary_loss_clip": 0.01189325,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.05700314,
      "balance_loss_mlp": 1.0180757,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.500190845471905,
      "language_loss": 0.85392058,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87608737,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.6359174251556396
    },
    {
      "auxiliary_loss_clip": 0.01195949,
      "auxiliary_loss_mlp": 0.01034201,
      "balance_loss_clip": 1.0612992,
      "balance_loss_mlp": 1.02496862,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.8871391558540662,
      "language_loss": 0.76562786,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78792942,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.637679100036621
    },
    {
      "auxiliary_loss_clip": 0.01210608,
      "auxiliary_loss_mlp": 0.01032121,
      "balance_loss_clip": 1.06260073,
      "balance_loss_mlp": 1.02245903,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.025722708563253,
      "language_loss": 0.7809267,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80335397,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.6851370334625244
    },
    {
      "auxiliary_loss_clip": 0.01206488,
      "auxiliary_loss_mlp": 0.00903173,
      "balance_loss_clip": 1.06170797,
      "balance_loss_mlp": 1.00213599,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 1.9843496547955661,
      "language_loss": 0.84503716,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.86613375,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.5877840518951416
    },
    {
      "auxiliary_loss_clip": 0.01207871,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.06228352,
      "balance_loss_mlp": 1.02642691,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.7373235376876226,
      "language_loss": 0.79339385,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81582773,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 2.6143486499786377
    },
    {
      "auxiliary_loss_clip": 0.01203328,
      "auxiliary_loss_mlp": 0.0090401,
      "balance_loss_clip": 1.05992651,
      "balance_loss_mlp": 1.00221968,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.2344682284572643,
      "language_loss": 0.80398691,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82506025,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.649259090423584
    },
    {
      "auxiliary_loss_clip": 0.01112906,
      "auxiliary_loss_mlp": 0.00892477,
      "balance_loss_clip": 1.03553343,
      "balance_loss_mlp": 1.00056386,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8172218919885297,
      "language_loss": 0.59386766,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61392152,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.0910704135894775
    },
    {
      "auxiliary_loss_clip": 0.01112434,
      "auxiliary_loss_mlp": 0.01007027,
      "balance_loss_clip": 1.03523636,
      "balance_loss_mlp": 1.00556111,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8084383432077498,
      "language_loss": 0.55777037,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57896495,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.1257131099700928
    },
    {
      "auxiliary_loss_clip": 0.01195294,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.05779028,
      "balance_loss_mlp": 1.01680124,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.8633208189340422,
      "language_loss": 0.82457769,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.84678596,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.6599605083465576
    },
    {
      "auxiliary_loss_clip": 0.01194868,
      "auxiliary_loss_mlp": 0.01030805,
      "balance_loss_clip": 1.05796683,
      "balance_loss_mlp": 1.02077329,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.6397662245239335,
      "language_loss": 0.86318016,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.88543689,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.5990512371063232
    },
    {
      "auxiliary_loss_clip": 0.0119419,
      "auxiliary_loss_mlp": 0.01036619,
      "balance_loss_clip": 1.0590961,
      "balance_loss_mlp": 1.02757108,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.5715235592712473,
      "language_loss": 0.75835842,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.78066653,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.732815980911255
    },
    {
      "auxiliary_loss_clip": 0.01169604,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.05267692,
      "balance_loss_mlp": 1.02026212,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 2.438609669057977,
      "language_loss": 0.75927818,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78126991,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.762087821960449
    },
    {
      "auxiliary_loss_clip": 0.01157773,
      "auxiliary_loss_mlp": 0.01038828,
      "balance_loss_clip": 1.04935026,
      "balance_loss_mlp": 1.0291003,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 1.8426061745141162,
      "language_loss": 0.86966026,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89162624,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.7623586654663086
    },
    {
      "auxiliary_loss_clip": 0.0118564,
      "auxiliary_loss_mlp": 0.01034309,
      "balance_loss_clip": 1.05918157,
      "balance_loss_mlp": 1.02508259,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 3.02366974588238,
      "language_loss": 0.89002013,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.91221958,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 3.654409646987915
    },
    {
      "auxiliary_loss_clip": 0.01207716,
      "auxiliary_loss_mlp": 0.01039549,
      "balance_loss_clip": 1.06202388,
      "balance_loss_mlp": 1.03049469,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.9914711196901798,
      "language_loss": 0.74000227,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76247489,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.6392533779144287
    },
    {
      "auxiliary_loss_clip": 0.01174825,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.05528939,
      "balance_loss_mlp": 1.01946342,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 5.5410746294112565,
      "language_loss": 0.75208688,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.77412343,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 3.524521827697754
    },
    {
      "auxiliary_loss_clip": 0.01195614,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.06009328,
      "balance_loss_mlp": 1.02148557,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.6518625981641817,
      "language_loss": 0.87906671,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90132296,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 3.6759109497070312
    },
    {
      "auxiliary_loss_clip": 0.01187985,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.05639005,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.6814901431979479,
      "language_loss": 0.69605368,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71819663,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.7206015586853027
    },
    {
      "auxiliary_loss_clip": 0.01189397,
      "auxiliary_loss_mlp": 0.01034057,
      "balance_loss_clip": 1.05955184,
      "balance_loss_mlp": 1.02444887,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.265617980602799,
      "language_loss": 0.829198,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85143256,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.6333298683166504
    },
    {
      "auxiliary_loss_clip": 0.01155159,
      "auxiliary_loss_mlp": 0.01041457,
      "balance_loss_clip": 1.05235553,
      "balance_loss_mlp": 1.0322721,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 1.960823242327866,
      "language_loss": 0.75693828,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.77890444,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.72684383392334
    },
    {
      "auxiliary_loss_clip": 0.01207687,
      "auxiliary_loss_mlp": 0.01037328,
      "balance_loss_clip": 1.05953026,
      "balance_loss_mlp": 1.0269804,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 2.595148773929596,
      "language_loss": 0.8372488,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85969889,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.6139206886291504
    },
    {
      "auxiliary_loss_clip": 0.01203739,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.05897307,
      "balance_loss_mlp": 1.02788162,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.4900484445722646,
      "language_loss": 0.78882492,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81124365,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 3.5442311763763428
    },
    {
      "auxiliary_loss_clip": 0.01198433,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.05693281,
      "balance_loss_mlp": 1.02026463,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 1.9058371829507654,
      "language_loss": 0.81623036,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83851302,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.6550989151000977
    },
    {
      "auxiliary_loss_clip": 0.01205267,
      "auxiliary_loss_mlp": 0.01031903,
      "balance_loss_clip": 1.06120276,
      "balance_loss_mlp": 1.02269399,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.7601940354124397,
      "language_loss": 0.76074439,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78311604,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.6802074909210205
    },
    {
      "auxiliary_loss_clip": 0.01181952,
      "auxiliary_loss_mlp": 0.00905314,
      "balance_loss_clip": 1.06011164,
      "balance_loss_mlp": 1.00215554,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.7292333065845717,
      "language_loss": 0.72883058,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.74970329,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.667496681213379
    },
    {
      "auxiliary_loss_clip": 0.0118658,
      "auxiliary_loss_mlp": 0.01035537,
      "balance_loss_clip": 1.0601697,
      "balance_loss_mlp": 1.0260005,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.802363381211574,
      "language_loss": 0.73765922,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.7598803,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.9288763999938965
    },
    {
      "auxiliary_loss_clip": 0.01171823,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.05381536,
      "balance_loss_mlp": 1.02550006,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 2.0690889487436315,
      "language_loss": 0.75506067,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77712762,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.6636438369750977
    },
    {
      "auxiliary_loss_clip": 0.01184657,
      "auxiliary_loss_mlp": 0.01035744,
      "balance_loss_clip": 1.05934715,
      "balance_loss_mlp": 1.02632642,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 3.0389912779545214,
      "language_loss": 0.66976333,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69196725,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.916888952255249
    },
    {
      "auxiliary_loss_clip": 0.01178378,
      "auxiliary_loss_mlp": 0.01031553,
      "balance_loss_clip": 1.0527885,
      "balance_loss_mlp": 1.02229619,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 4.000449402605563,
      "language_loss": 0.8277185,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.84981775,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.722391366958618
    },
    {
      "auxiliary_loss_clip": 0.01095608,
      "auxiliary_loss_mlp": 0.01004944,
      "balance_loss_clip": 1.02546895,
      "balance_loss_mlp": 1.00337064,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7741341725545657,
      "language_loss": 0.52568185,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54668742,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.2724897861480713
    },
    {
      "auxiliary_loss_clip": 0.0116572,
      "auxiliary_loss_mlp": 0.01035382,
      "balance_loss_clip": 1.05645502,
      "balance_loss_mlp": 1.02705491,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.2312253588625928,
      "language_loss": 0.95826352,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98027456,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.724327802658081
    },
    {
      "auxiliary_loss_clip": 0.01197367,
      "auxiliary_loss_mlp": 0.01030083,
      "balance_loss_clip": 1.06320882,
      "balance_loss_mlp": 1.02038515,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 2.9784915454639473,
      "language_loss": 0.8302511,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85252559,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.7506649494171143
    },
    {
      "auxiliary_loss_clip": 0.0120492,
      "auxiliary_loss_mlp": 0.01029171,
      "balance_loss_clip": 1.06076455,
      "balance_loss_mlp": 1.01933026,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 2.4369966952920135,
      "language_loss": 0.76827753,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79061842,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.677081346511841
    },
    {
      "auxiliary_loss_clip": 0.01089762,
      "auxiliary_loss_mlp": 0.0100473,
      "balance_loss_clip": 1.0275147,
      "balance_loss_mlp": 1.0031569,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8830626226280295,
      "language_loss": 0.60519081,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62613577,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.202178478240967
    },
    {
      "auxiliary_loss_clip": 0.01187621,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.05539525,
      "balance_loss_mlp": 1.01632822,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 2.2395795108021157,
      "language_loss": 0.79079163,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.8129245,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.7993292808532715
    },
    {
      "auxiliary_loss_clip": 0.01163251,
      "auxiliary_loss_mlp": 0.01031376,
      "balance_loss_clip": 1.05430603,
      "balance_loss_mlp": 1.02226806,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 2.431504137042853,
      "language_loss": 0.68318951,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70513576,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.8262665271759033
    },
    {
      "auxiliary_loss_clip": 0.01164966,
      "auxiliary_loss_mlp": 0.01031236,
      "balance_loss_clip": 1.05138326,
      "balance_loss_mlp": 1.02262878,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.8805255629707225,
      "language_loss": 0.68008316,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.70204508,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.757401466369629
    },
    {
      "auxiliary_loss_clip": 0.01178091,
      "auxiliary_loss_mlp": 0.01034461,
      "balance_loss_clip": 1.05263186,
      "balance_loss_mlp": 1.02481139,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 1.8217722192537347,
      "language_loss": 0.83549619,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.85762173,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.641180992126465
    },
    {
      "auxiliary_loss_clip": 0.01174939,
      "auxiliary_loss_mlp": 0.01034281,
      "balance_loss_clip": 1.05430853,
      "balance_loss_mlp": 1.02556133,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 2.1110904436126714,
      "language_loss": 0.77528667,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79737884,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.687626838684082
    },
    {
      "auxiliary_loss_clip": 0.0109197,
      "auxiliary_loss_mlp": 0.00892816,
      "balance_loss_clip": 1.02349818,
      "balance_loss_mlp": 1.00050724,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8036082990966362,
      "language_loss": 0.6038574,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62370527,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.2787668704986572
    },
    {
      "auxiliary_loss_clip": 0.01162331,
      "auxiliary_loss_mlp": 0.01033586,
      "balance_loss_clip": 1.05188107,
      "balance_loss_mlp": 1.02499092,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.6584557254826342,
      "language_loss": 0.78853232,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.8104915,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 3.7019717693328857
    },
    {
      "auxiliary_loss_clip": 0.01179642,
      "auxiliary_loss_mlp": 0.01026482,
      "balance_loss_clip": 1.05822706,
      "balance_loss_mlp": 1.01762497,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 1.814355118871195,
      "language_loss": 0.77665216,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.79871339,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.7455601692199707
    },
    {
      "auxiliary_loss_clip": 0.01165009,
      "auxiliary_loss_mlp": 0.01028892,
      "balance_loss_clip": 1.05251527,
      "balance_loss_mlp": 1.01951027,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.1967597161061065,
      "language_loss": 0.92308015,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94501913,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 3.7151546478271484
    },
    {
      "auxiliary_loss_clip": 0.01098743,
      "auxiliary_loss_mlp": 0.01001374,
      "balance_loss_clip": 1.02201843,
      "balance_loss_mlp": 0.99985391,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8906975065883561,
      "language_loss": 0.63557833,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65657949,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 4.038036108016968
    },
    {
      "auxiliary_loss_clip": 0.01207177,
      "auxiliary_loss_mlp": 0.01034315,
      "balance_loss_clip": 1.06271434,
      "balance_loss_mlp": 1.02427721,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 2.5404360991557446,
      "language_loss": 0.80830979,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.83072472,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.663984775543213
    },
    {
      "auxiliary_loss_clip": 0.01201705,
      "auxiliary_loss_mlp": 0.01031906,
      "balance_loss_clip": 1.06068182,
      "balance_loss_mlp": 1.02297103,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 2.936588608866711,
      "language_loss": 0.87186623,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89420235,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.647557497024536
    },
    {
      "auxiliary_loss_clip": 0.01159712,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.05240679,
      "balance_loss_mlp": 1.02288604,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.7631556736817355,
      "language_loss": 0.77512908,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79704136,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.8165102005004883
    },
    {
      "auxiliary_loss_clip": 0.01203233,
      "auxiliary_loss_mlp": 0.01030513,
      "balance_loss_clip": 1.05993664,
      "balance_loss_mlp": 1.02242506,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 2.0351786020148617,
      "language_loss": 0.77575469,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79809213,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.654703378677368
    },
    {
      "auxiliary_loss_clip": 0.01174434,
      "auxiliary_loss_mlp": 0.00904007,
      "balance_loss_clip": 1.0541147,
      "balance_loss_mlp": 1.00230694,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.9179481039675577,
      "language_loss": 0.84340942,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.8641938,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 3.557621955871582
    },
    {
      "auxiliary_loss_clip": 0.01191464,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 1.05543721,
      "balance_loss_mlp": 1.02161527,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 2.0336646907075258,
      "language_loss": 0.66380608,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.6860261,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.766775131225586
    },
    {
      "auxiliary_loss_clip": 0.01193354,
      "auxiliary_loss_mlp": 0.01029686,
      "balance_loss_clip": 1.05987859,
      "balance_loss_mlp": 1.0207988,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.710584750457759,
      "language_loss": 0.89281523,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91504562,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.6316611766815186
    },
    {
      "auxiliary_loss_clip": 0.01199369,
      "auxiliary_loss_mlp": 0.01039764,
      "balance_loss_clip": 1.05989397,
      "balance_loss_mlp": 1.03023386,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.6714134569209262,
      "language_loss": 0.73107207,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75346345,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.61979603767395
    },
    {
      "auxiliary_loss_clip": 0.01187242,
      "auxiliary_loss_mlp": 0.01039721,
      "balance_loss_clip": 1.05609584,
      "balance_loss_mlp": 1.03074455,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.9589884847531172,
      "language_loss": 0.85255075,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87482035,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.697053909301758
    },
    {
      "auxiliary_loss_clip": 0.01193197,
      "auxiliary_loss_mlp": 0.01032876,
      "balance_loss_clip": 1.05720437,
      "balance_loss_mlp": 1.02348781,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.017578424338764,
      "language_loss": 0.85767365,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87993431,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.7056632041931152
    },
    {
      "auxiliary_loss_clip": 0.01193606,
      "auxiliary_loss_mlp": 0.00903334,
      "balance_loss_clip": 1.05728364,
      "balance_loss_mlp": 1.00232697,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.7926689650322778,
      "language_loss": 0.85705614,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.87802553,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.69608998298645
    },
    {
      "auxiliary_loss_clip": 0.01202308,
      "auxiliary_loss_mlp": 0.01031699,
      "balance_loss_clip": 1.05927229,
      "balance_loss_mlp": 1.02321708,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.6241267438160936,
      "language_loss": 0.84487426,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86721432,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.6036903858184814
    },
    {
      "auxiliary_loss_clip": 0.0120156,
      "auxiliary_loss_mlp": 0.01033219,
      "balance_loss_clip": 1.05985224,
      "balance_loss_mlp": 1.0239861,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 2.196871583849105,
      "language_loss": 0.71210378,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73445165,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.6692330837249756
    },
    {
      "auxiliary_loss_clip": 0.01177369,
      "auxiliary_loss_mlp": 0.01032328,
      "balance_loss_clip": 1.0582881,
      "balance_loss_mlp": 1.02363813,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.7599534299369803,
      "language_loss": 0.81821382,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84031069,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.6230180263519287
    },
    {
      "auxiliary_loss_clip": 0.01181597,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.05649233,
      "balance_loss_mlp": 1.02249956,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.7402863828208899,
      "language_loss": 0.77422231,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79634893,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.722877264022827
    },
    {
      "auxiliary_loss_clip": 0.01189784,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.05524302,
      "balance_loss_mlp": 1.02323294,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.7361737836862283,
      "language_loss": 0.83185065,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.8540684,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.6785871982574463
    },
    {
      "auxiliary_loss_clip": 0.01097129,
      "auxiliary_loss_mlp": 0.01004698,
      "balance_loss_clip": 1.02048564,
      "balance_loss_mlp": 1.00308895,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7695510113797798,
      "language_loss": 0.55555558,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57657385,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.2146811485290527
    },
    {
      "auxiliary_loss_clip": 0.01200794,
      "auxiliary_loss_mlp": 0.01031421,
      "balance_loss_clip": 1.06028175,
      "balance_loss_mlp": 1.02179515,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 2.5016266899062125,
      "language_loss": 0.82095891,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84328109,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.6147985458374023
    },
    {
      "auxiliary_loss_clip": 0.01178991,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.05642867,
      "balance_loss_mlp": 1.02067804,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.8057217146261102,
      "language_loss": 0.8407433,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86282539,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.7261128425598145
    },
    {
      "auxiliary_loss_clip": 0.01176264,
      "auxiliary_loss_mlp": 0.01031289,
      "balance_loss_clip": 1.05703139,
      "balance_loss_mlp": 1.02241409,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.4375803264212927,
      "language_loss": 0.71195066,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73402619,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.6918833255767822
    },
    {
      "auxiliary_loss_clip": 0.01195636,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.06109905,
      "balance_loss_mlp": 1.01929772,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 2.7031263738310796,
      "language_loss": 0.76711869,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.78935492,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.6092257499694824
    },
    {
      "auxiliary_loss_clip": 0.01192062,
      "auxiliary_loss_mlp": 0.01031107,
      "balance_loss_clip": 1.0578413,
      "balance_loss_mlp": 1.02177334,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 2.3975841788887458,
      "language_loss": 0.73111236,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.75334406,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.6054000854492188
    },
    {
      "auxiliary_loss_clip": 0.0118407,
      "auxiliary_loss_mlp": 0.01037493,
      "balance_loss_clip": 1.0555737,
      "balance_loss_mlp": 1.02812362,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 2.1986771583574143,
      "language_loss": 0.88330793,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90552354,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 3.591076135635376
    },
    {
      "auxiliary_loss_clip": 0.011798,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.05294359,
      "balance_loss_mlp": 1.0259198,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.579639081628879,
      "language_loss": 0.7746377,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79678774,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.763509750366211
    },
    {
      "auxiliary_loss_clip": 0.01163056,
      "auxiliary_loss_mlp": 0.01034805,
      "balance_loss_clip": 1.05328989,
      "balance_loss_mlp": 1.02583432,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.0863849105325585,
      "language_loss": 0.64126337,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66324204,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.6415090560913086
    },
    {
      "auxiliary_loss_clip": 0.01097628,
      "auxiliary_loss_mlp": 0.01001771,
      "balance_loss_clip": 1.02104163,
      "balance_loss_mlp": 1.00013781,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7751882699891108,
      "language_loss": 0.56895351,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.58994746,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 4.946678161621094
    },
    {
      "auxiliary_loss_clip": 0.01202876,
      "auxiliary_loss_mlp": 0.01029879,
      "balance_loss_clip": 1.05936742,
      "balance_loss_mlp": 1.02086091,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.180420728809363,
      "language_loss": 0.72512817,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74745578,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.594191789627075
    },
    {
      "auxiliary_loss_clip": 0.01190239,
      "auxiliary_loss_mlp": 0.01025667,
      "balance_loss_clip": 1.05510116,
      "balance_loss_mlp": 1.01676846,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.7089572579494239,
      "language_loss": 0.69669819,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.71885729,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.6453659534454346
    },
    {
      "auxiliary_loss_clip": 0.01188844,
      "auxiliary_loss_mlp": 0.01030157,
      "balance_loss_clip": 1.05847454,
      "balance_loss_mlp": 1.02121067,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 2.016146020381793,
      "language_loss": 0.88396966,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90615964,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.7010560035705566
    },
    {
      "auxiliary_loss_clip": 0.0118303,
      "auxiliary_loss_mlp": 0.01029261,
      "balance_loss_clip": 1.05705738,
      "balance_loss_mlp": 1.02023733,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 1.9970051840068588,
      "language_loss": 0.79536402,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81748694,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.6663389205932617
    },
    {
      "auxiliary_loss_clip": 0.01194671,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05737233,
      "balance_loss_mlp": 1.0219605,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.697970482657548,
      "language_loss": 0.7209608,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.74322248,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 3.5603668689727783
    },
    {
      "auxiliary_loss_clip": 0.01165603,
      "auxiliary_loss_mlp": 0.01033411,
      "balance_loss_clip": 1.05081379,
      "balance_loss_mlp": 1.02416587,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 3.436656881095995,
      "language_loss": 0.9006815,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92267162,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.695754051208496
    },
    {
      "auxiliary_loss_clip": 0.01098578,
      "auxiliary_loss_mlp": 0.01001917,
      "balance_loss_clip": 1.04045606,
      "balance_loss_mlp": 0.99999774,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7913834773921927,
      "language_loss": 0.6285609,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64956594,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.2731850147247314
    },
    {
      "auxiliary_loss_clip": 0.01175022,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05417228,
      "balance_loss_mlp": 1.02347541,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.750857877272573,
      "language_loss": 0.81905025,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.84113073,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.7081336975097656
    },
    {
      "auxiliary_loss_clip": 0.01205861,
      "auxiliary_loss_mlp": 0.01036594,
      "balance_loss_clip": 1.06203377,
      "balance_loss_mlp": 1.02782011,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 2.1941273928052634,
      "language_loss": 0.75013745,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77256203,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.55509877204895
    },
    {
      "auxiliary_loss_clip": 0.01192936,
      "auxiliary_loss_mlp": 0.01030057,
      "balance_loss_clip": 1.05958033,
      "balance_loss_mlp": 1.02133131,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.178485830498143,
      "language_loss": 0.8172667,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83949661,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.6454625129699707
    },
    {
      "auxiliary_loss_clip": 0.01157928,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.04610407,
      "balance_loss_mlp": 1.02235377,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 4.634196736136443,
      "language_loss": 0.7437278,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76562446,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.6844735145568848
    },
    {
      "auxiliary_loss_clip": 0.01161412,
      "auxiliary_loss_mlp": 0.01034979,
      "balance_loss_clip": 1.05505252,
      "balance_loss_mlp": 1.02544177,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.975004229495652,
      "language_loss": 0.80214101,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82410491,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.7136240005493164
    },
    {
      "auxiliary_loss_clip": 0.01169619,
      "auxiliary_loss_mlp": 0.01029214,
      "balance_loss_clip": 1.05298185,
      "balance_loss_mlp": 1.02005267,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.245757853716316,
      "language_loss": 0.7790904,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80107868,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 2.6997125148773193
    },
    {
      "auxiliary_loss_clip": 0.01173703,
      "auxiliary_loss_mlp": 0.01035877,
      "balance_loss_clip": 1.05287457,
      "balance_loss_mlp": 1.02631676,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.8672593542909275,
      "language_loss": 0.73299688,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75509274,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.658351182937622
    },
    {
      "auxiliary_loss_clip": 0.01194087,
      "auxiliary_loss_mlp": 0.01036121,
      "balance_loss_clip": 1.06349397,
      "balance_loss_mlp": 1.02727008,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 2.785441720636355,
      "language_loss": 0.75815463,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.78045672,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.752864360809326
    },
    {
      "auxiliary_loss_clip": 0.01171333,
      "auxiliary_loss_mlp": 0.01028076,
      "balance_loss_clip": 1.05589318,
      "balance_loss_mlp": 1.01903439,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 1.896238390247391,
      "language_loss": 0.8057363,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82773042,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.653203248977661
    },
    {
      "auxiliary_loss_clip": 0.01198723,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.06085491,
      "balance_loss_mlp": 1.01907277,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 2.153257850593795,
      "language_loss": 0.70456481,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.7268337,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.640404224395752
    },
    {
      "auxiliary_loss_clip": 0.01190923,
      "auxiliary_loss_mlp": 0.01033551,
      "balance_loss_clip": 1.05857539,
      "balance_loss_mlp": 1.02408576,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 2.2730471957622593,
      "language_loss": 0.75895721,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78120196,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.6130378246307373
    },
    {
      "auxiliary_loss_clip": 0.0118628,
      "auxiliary_loss_mlp": 0.00902926,
      "balance_loss_clip": 1.05488908,
      "balance_loss_mlp": 1.00210953,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.1093146436407446,
      "language_loss": 0.84429574,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86518782,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.655095338821411
    },
    {
      "auxiliary_loss_clip": 0.01187207,
      "auxiliary_loss_mlp": 0.01029958,
      "balance_loss_clip": 1.05195427,
      "balance_loss_mlp": 1.0217979,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.157867813963304,
      "language_loss": 0.91865432,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.94082594,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.5964393615722656
    },
    {
      "auxiliary_loss_clip": 0.01156749,
      "auxiliary_loss_mlp": 0.01037002,
      "balance_loss_clip": 1.05249929,
      "balance_loss_mlp": 1.02741706,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.483720057659172,
      "language_loss": 0.82679468,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84873223,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.7667126655578613
    },
    {
      "auxiliary_loss_clip": 0.01186602,
      "auxiliary_loss_mlp": 0.01036337,
      "balance_loss_clip": 1.05480516,
      "balance_loss_mlp": 1.02789116,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.645314621027112,
      "language_loss": 0.71344936,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73567867,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.7524654865264893
    },
    {
      "auxiliary_loss_clip": 0.01195895,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.05646062,
      "balance_loss_mlp": 1.02313352,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.9601342297110371,
      "language_loss": 0.80263221,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82491207,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.6359355449676514
    },
    {
      "auxiliary_loss_clip": 0.01200363,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.0589745,
      "balance_loss_mlp": 1.02446508,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 2.359937190169691,
      "language_loss": 0.75585639,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77819461,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 3.470925807952881
    },
    {
      "auxiliary_loss_clip": 0.01193627,
      "auxiliary_loss_mlp": 0.01037515,
      "balance_loss_clip": 1.05763626,
      "balance_loss_mlp": 1.02794814,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.8657733327598278,
      "language_loss": 0.7953155,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81762695,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 2.844461441040039
    },
    {
      "auxiliary_loss_clip": 0.01099824,
      "auxiliary_loss_mlp": 0.01002728,
      "balance_loss_clip": 1.02344024,
      "balance_loss_mlp": 1.00114274,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.767103882227327,
      "language_loss": 0.55249763,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57352316,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 4.169220685958862
    },
    {
      "auxiliary_loss_clip": 0.01177074,
      "auxiliary_loss_mlp": 0.01034801,
      "balance_loss_clip": 1.05359435,
      "balance_loss_mlp": 1.02534735,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.963340938527898,
      "language_loss": 0.80181694,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82393569,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 3.627147674560547
    },
    {
      "auxiliary_loss_clip": 0.01199484,
      "auxiliary_loss_mlp": 0.01029454,
      "balance_loss_clip": 1.05805397,
      "balance_loss_mlp": 1.02047741,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.621055091838063,
      "language_loss": 0.83861583,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86090523,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.599487066268921
    },
    {
      "auxiliary_loss_clip": 0.01200675,
      "auxiliary_loss_mlp": 0.01029224,
      "balance_loss_clip": 1.05866838,
      "balance_loss_mlp": 1.02039683,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 1.8770155056265898,
      "language_loss": 0.84277761,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86507666,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.6072020530700684
    },
    {
      "auxiliary_loss_clip": 0.01157754,
      "auxiliary_loss_mlp": 0.01037393,
      "balance_loss_clip": 1.05450213,
      "balance_loss_mlp": 1.02830362,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.7299558057553088,
      "language_loss": 0.86436558,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88631707,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.7046549320220947
    },
    {
      "auxiliary_loss_clip": 0.0117849,
      "auxiliary_loss_mlp": 0.01034708,
      "balance_loss_clip": 1.06308532,
      "balance_loss_mlp": 1.02555907,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 5.1524646507664125,
      "language_loss": 0.72971386,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75184584,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 3.733020782470703
    },
    {
      "auxiliary_loss_clip": 0.01200211,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.05645621,
      "balance_loss_mlp": 1.01953387,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 2.4013869287754606,
      "language_loss": 0.8471508,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.8694306,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.6037163734436035
    },
    {
      "auxiliary_loss_clip": 0.01190045,
      "auxiliary_loss_mlp": 0.01035895,
      "balance_loss_clip": 1.05736113,
      "balance_loss_mlp": 1.02614975,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 12.715875353227107,
      "language_loss": 0.79707772,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81933713,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.6365909576416016
    },
    {
      "auxiliary_loss_clip": 0.01184085,
      "auxiliary_loss_mlp": 0.01030362,
      "balance_loss_clip": 1.05503035,
      "balance_loss_mlp": 1.02129006,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 1.7919574248890051,
      "language_loss": 0.66183031,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68397474,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.7153878211975098
    },
    {
      "auxiliary_loss_clip": 0.0118752,
      "auxiliary_loss_mlp": 0.01033487,
      "balance_loss_clip": 1.05639517,
      "balance_loss_mlp": 1.02384925,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 2.0609931495655434,
      "language_loss": 0.85019588,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87240601,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.72963285446167
    },
    {
      "auxiliary_loss_clip": 0.01187625,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.0530498,
      "balance_loss_mlp": 1.01941407,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.209982501579194,
      "language_loss": 0.86706173,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88921821,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.6606180667877197
    },
    {
      "auxiliary_loss_clip": 0.01177058,
      "auxiliary_loss_mlp": 0.01031454,
      "balance_loss_clip": 1.05564916,
      "balance_loss_mlp": 1.02284706,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.7987797010922852,
      "language_loss": 0.75639462,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.7784797,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.6575472354888916
    },
    {
      "auxiliary_loss_clip": 0.01189884,
      "auxiliary_loss_mlp": 0.0103173,
      "balance_loss_clip": 1.05653429,
      "balance_loss_mlp": 1.02293253,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.5415686475515047,
      "language_loss": 0.69476843,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71698463,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.8748791217803955
    },
    {
      "auxiliary_loss_clip": 0.01187973,
      "auxiliary_loss_mlp": 0.01030952,
      "balance_loss_clip": 1.05485821,
      "balance_loss_mlp": 1.02288103,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.8268077996760421,
      "language_loss": 0.81004632,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.83223563,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.6137189865112305
    },
    {
      "auxiliary_loss_clip": 0.0118373,
      "auxiliary_loss_mlp": 0.01037072,
      "balance_loss_clip": 1.05603766,
      "balance_loss_mlp": 1.02822113,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.708415693243243,
      "language_loss": 0.71677357,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.7389816,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 2.6779603958129883
    },
    {
      "auxiliary_loss_clip": 0.01179096,
      "auxiliary_loss_mlp": 0.01034876,
      "balance_loss_clip": 1.05938423,
      "balance_loss_mlp": 1.02623892,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 2.196363203565145,
      "language_loss": 0.74418843,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.7663281,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.6566824913024902
    },
    {
      "auxiliary_loss_clip": 0.01203023,
      "auxiliary_loss_mlp": 0.01031711,
      "balance_loss_clip": 1.05687845,
      "balance_loss_mlp": 1.02226329,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.634148529245855,
      "language_loss": 0.9155724,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93791968,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.5394957065582275
    },
    {
      "auxiliary_loss_clip": 0.01204182,
      "auxiliary_loss_mlp": 0.01034629,
      "balance_loss_clip": 1.05836225,
      "balance_loss_mlp": 1.02491379,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.5659641304610257,
      "language_loss": 0.81291163,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83529979,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.5828702449798584
    },
    {
      "auxiliary_loss_clip": 0.01165742,
      "auxiliary_loss_mlp": 0.0103484,
      "balance_loss_clip": 1.05674672,
      "balance_loss_mlp": 1.02644789,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.6725170392560686,
      "language_loss": 0.87216687,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89417273,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.688553810119629
    },
    {
      "auxiliary_loss_clip": 0.01183442,
      "auxiliary_loss_mlp": 0.01038129,
      "balance_loss_clip": 1.0537827,
      "balance_loss_mlp": 1.0282886,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 2.05426484581957,
      "language_loss": 0.85271966,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87493539,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.6335625648498535
    },
    {
      "auxiliary_loss_clip": 0.01189952,
      "auxiliary_loss_mlp": 0.01025738,
      "balance_loss_clip": 1.05785322,
      "balance_loss_mlp": 1.01668406,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 2.781111548261414,
      "language_loss": 0.61740863,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63956559,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.6593925952911377
    },
    {
      "auxiliary_loss_clip": 0.01156918,
      "auxiliary_loss_mlp": 0.01032664,
      "balance_loss_clip": 1.05138147,
      "balance_loss_mlp": 1.02347302,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 2.02384305068478,
      "language_loss": 0.73133957,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.7532354,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.842111825942993
    },
    {
      "auxiliary_loss_clip": 0.01182892,
      "auxiliary_loss_mlp": 0.01028567,
      "balance_loss_clip": 1.05530477,
      "balance_loss_mlp": 1.02013302,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 1.9804082312650224,
      "language_loss": 0.84711921,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86923385,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.7129414081573486
    },
    {
      "auxiliary_loss_clip": 0.01197171,
      "auxiliary_loss_mlp": 0.01035182,
      "balance_loss_clip": 1.05613947,
      "balance_loss_mlp": 1.02612782,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.745290316630873,
      "language_loss": 0.72152746,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.74385095,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.6304750442504883
    },
    {
      "auxiliary_loss_clip": 0.01200095,
      "auxiliary_loss_mlp": 0.01032217,
      "balance_loss_clip": 1.05565786,
      "balance_loss_mlp": 1.02297211,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.6268663024697845,
      "language_loss": 0.78860211,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.81092525,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 3.45729660987854
    },
    {
      "auxiliary_loss_clip": 0.01191018,
      "auxiliary_loss_mlp": 0.01041455,
      "balance_loss_clip": 1.0583775,
      "balance_loss_mlp": 1.03244913,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.233511378341585,
      "language_loss": 0.84212995,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86445463,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.67441725730896
    },
    {
      "auxiliary_loss_clip": 0.01202478,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.05924511,
      "balance_loss_mlp": 1.01937926,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.7983913697275793,
      "language_loss": 0.74722642,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76953995,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.621342897415161
    },
    {
      "auxiliary_loss_clip": 0.01184023,
      "auxiliary_loss_mlp": 0.01033323,
      "balance_loss_clip": 1.05635929,
      "balance_loss_mlp": 1.02425086,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 15.846524033190406,
      "language_loss": 0.85981381,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88198733,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 3.652323007583618
    },
    {
      "auxiliary_loss_clip": 0.0115793,
      "auxiliary_loss_mlp": 0.0102954,
      "balance_loss_clip": 1.05388021,
      "balance_loss_mlp": 1.02092087,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 2.052355192845432,
      "language_loss": 0.77472126,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.79659593,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 3.737743854522705
    },
    {
      "auxiliary_loss_clip": 0.01194115,
      "auxiliary_loss_mlp": 0.01030645,
      "balance_loss_clip": 1.05992353,
      "balance_loss_mlp": 1.02156115,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.0927270189403444,
      "language_loss": 0.84143567,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86368322,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.652639389038086
    },
    {
      "auxiliary_loss_clip": 0.01097144,
      "auxiliary_loss_mlp": 0.01010577,
      "balance_loss_clip": 1.02686679,
      "balance_loss_mlp": 1.00912857,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6473785705768252,
      "language_loss": 0.5025475,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52362466,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.204672336578369
    },
    {
      "auxiliary_loss_clip": 0.01199567,
      "auxiliary_loss_mlp": 0.01034912,
      "balance_loss_clip": 1.05854988,
      "balance_loss_mlp": 1.0258162,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 2.0757264709779393,
      "language_loss": 0.78973544,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.81208026,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.598145008087158
    },
    {
      "auxiliary_loss_clip": 0.01190395,
      "auxiliary_loss_mlp": 0.01036137,
      "balance_loss_clip": 1.05600166,
      "balance_loss_mlp": 1.02647495,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.721697594132666,
      "language_loss": 0.81695372,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83921909,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 3.617920398712158
    },
    {
      "auxiliary_loss_clip": 0.01190795,
      "auxiliary_loss_mlp": 0.01029608,
      "balance_loss_clip": 1.05848384,
      "balance_loss_mlp": 1.02059603,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.5362509163889755,
      "language_loss": 0.80286717,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82507128,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.649121046066284
    },
    {
      "auxiliary_loss_clip": 0.01202474,
      "auxiliary_loss_mlp": 0.0090322,
      "balance_loss_clip": 1.05945659,
      "balance_loss_mlp": 1.0019722,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 2.1164323903409796,
      "language_loss": 0.8692987,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.89035559,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.5906918048858643
    },
    {
      "auxiliary_loss_clip": 0.0109636,
      "auxiliary_loss_mlp": 0.01002533,
      "balance_loss_clip": 1.02648175,
      "balance_loss_mlp": 1.00108457,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6691025720990773,
      "language_loss": 0.55669308,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.5776819,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.2517192363739014
    },
    {
      "auxiliary_loss_clip": 0.0119386,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 1.05491352,
      "balance_loss_mlp": 1.02139103,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.4584841505089363,
      "language_loss": 0.88162577,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.9038673,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.666130781173706
    },
    {
      "auxiliary_loss_clip": 0.01163718,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.04891253,
      "balance_loss_mlp": 1.0203383,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 2.051932525644631,
      "language_loss": 0.77221096,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79414886,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.704612970352173
    },
    {
      "auxiliary_loss_clip": 0.01201353,
      "auxiliary_loss_mlp": 0.01031426,
      "balance_loss_clip": 1.06029058,
      "balance_loss_mlp": 1.02299798,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.6167589817867192,
      "language_loss": 0.71996003,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74228781,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.6830363273620605
    },
    {
      "auxiliary_loss_clip": 0.01198676,
      "auxiliary_loss_mlp": 0.01028672,
      "balance_loss_clip": 1.05839479,
      "balance_loss_mlp": 1.02074432,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.6556089040910977,
      "language_loss": 0.87836707,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.90064055,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.580355405807495
    },
    {
      "auxiliary_loss_clip": 0.01185999,
      "auxiliary_loss_mlp": 0.01035331,
      "balance_loss_clip": 1.06035924,
      "balance_loss_mlp": 1.02607489,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 1.825248334569387,
      "language_loss": 0.75376296,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.7759763,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.7006869316101074
    },
    {
      "auxiliary_loss_clip": 0.01154468,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.04994273,
      "balance_loss_mlp": 1.01975822,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.870368348992066,
      "language_loss": 0.8535279,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87535501,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 2.6984965801239014
    },
    {
      "auxiliary_loss_clip": 0.01179496,
      "auxiliary_loss_mlp": 0.00902918,
      "balance_loss_clip": 1.05531514,
      "balance_loss_mlp": 1.00203395,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 2.825414974828946,
      "language_loss": 0.80112058,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82194471,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.6985113620758057
    },
    {
      "auxiliary_loss_clip": 0.01092602,
      "auxiliary_loss_mlp": 0.01003589,
      "balance_loss_clip": 1.02389681,
      "balance_loss_mlp": 1.00215244,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8744763515327796,
      "language_loss": 0.57432878,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59529066,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.197953701019287
    },
    {
      "auxiliary_loss_clip": 0.01170618,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.0554446,
      "balance_loss_mlp": 1.02321005,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.602998139866728,
      "language_loss": 0.71405739,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73608023,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.7552402019500732
    },
    {
      "auxiliary_loss_clip": 0.01193251,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.0561235,
      "balance_loss_mlp": 1.02130461,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 2.013021120864324,
      "language_loss": 0.66903251,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69127095,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.722320795059204
    },
    {
      "auxiliary_loss_clip": 0.01165316,
      "auxiliary_loss_mlp": 0.01035934,
      "balance_loss_clip": 1.05011404,
      "balance_loss_mlp": 1.02716613,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.6628308290791132,
      "language_loss": 0.80066562,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82267815,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.7280335426330566
    },
    {
      "auxiliary_loss_clip": 0.01181483,
      "auxiliary_loss_mlp": 0.01028856,
      "balance_loss_clip": 1.05732059,
      "balance_loss_mlp": 1.02039218,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.7702172625599173,
      "language_loss": 0.75625771,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.77836108,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.838207244873047
    },
    {
      "auxiliary_loss_clip": 0.01194624,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.05530107,
      "balance_loss_mlp": 1.02356625,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.7972983225415766,
      "language_loss": 0.78469723,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80697453,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.655693769454956
    },
    {
      "auxiliary_loss_clip": 0.0117105,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.05187535,
      "balance_loss_mlp": 1.02314305,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 3.483567640641486,
      "language_loss": 0.61600327,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63804311,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.7409653663635254
    },
    {
      "auxiliary_loss_clip": 0.01167542,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04897857,
      "balance_loss_mlp": 1.02126384,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 1.9462982011575576,
      "language_loss": 0.81837344,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.84034163,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.706625461578369
    },
    {
      "auxiliary_loss_clip": 0.01176565,
      "auxiliary_loss_mlp": 0.01028215,
      "balance_loss_clip": 1.05320227,
      "balance_loss_mlp": 1.019894,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.7222679350757804,
      "language_loss": 0.69924152,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72128928,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 3.572166919708252
    },
    {
      "auxiliary_loss_clip": 0.01180266,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.0567801,
      "balance_loss_mlp": 1.02365232,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 1.9773171998247479,
      "language_loss": 0.84335959,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86548162,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.6885361671447754
    },
    {
      "auxiliary_loss_clip": 0.01173722,
      "auxiliary_loss_mlp": 0.00902113,
      "balance_loss_clip": 1.0567739,
      "balance_loss_mlp": 1.00207162,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.3735210679933123,
      "language_loss": 0.75921392,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.77997231,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 3.5921103954315186
    },
    {
      "auxiliary_loss_clip": 0.01181941,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.05339479,
      "balance_loss_mlp": 1.02206194,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.292817384242398,
      "language_loss": 0.78563565,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80776322,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 3.6180737018585205
    },
    {
      "auxiliary_loss_clip": 0.01177037,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.05205584,
      "balance_loss_mlp": 1.0202632,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 2.3446191607660696,
      "language_loss": 0.74562496,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76767963,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.8080484867095947
    },
    {
      "auxiliary_loss_clip": 0.01164551,
      "auxiliary_loss_mlp": 0.01027417,
      "balance_loss_clip": 1.049371,
      "balance_loss_mlp": 1.02003789,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 2.8910123156737364,
      "language_loss": 0.74420017,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76611984,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.7341160774230957
    },
    {
      "auxiliary_loss_clip": 0.01155634,
      "auxiliary_loss_mlp": 0.01031448,
      "balance_loss_clip": 1.05074692,
      "balance_loss_mlp": 1.02270997,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.621930479185862,
      "language_loss": 0.88421535,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.90608609,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.7867507934570312
    },
    {
      "auxiliary_loss_clip": 0.01187558,
      "auxiliary_loss_mlp": 0.01028304,
      "balance_loss_clip": 1.05637622,
      "balance_loss_mlp": 1.01883256,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 1.9693291959788035,
      "language_loss": 0.78714168,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.8093003,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 3.5665502548217773
    },
    {
      "auxiliary_loss_clip": 0.01185897,
      "auxiliary_loss_mlp": 0.00901867,
      "balance_loss_clip": 1.05342996,
      "balance_loss_mlp": 1.00190759,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.591737739629921,
      "language_loss": 0.80017602,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.82105362,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.6564507484436035
    },
    {
      "auxiliary_loss_clip": 0.01165227,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 1.05232346,
      "balance_loss_mlp": 1.02540874,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 4.106208296602296,
      "language_loss": 0.8121292,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83412951,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.903428792953491
    },
    {
      "auxiliary_loss_clip": 0.01189194,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.05472326,
      "balance_loss_mlp": 1.02157271,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 1.7534755889918225,
      "language_loss": 0.86652982,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88872993,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.6898574829101562
    },
    {
      "auxiliary_loss_clip": 0.01166075,
      "auxiliary_loss_mlp": 0.01031884,
      "balance_loss_clip": 1.05149055,
      "balance_loss_mlp": 1.02332449,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 4.645475176865255,
      "language_loss": 0.65992433,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.6819039,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.737476348876953
    },
    {
      "auxiliary_loss_clip": 0.01186044,
      "auxiliary_loss_mlp": 0.01033188,
      "balance_loss_clip": 1.05560052,
      "balance_loss_mlp": 1.02409863,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.543857267363683,
      "language_loss": 0.9127267,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93491906,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.616910219192505
    },
    {
      "auxiliary_loss_clip": 0.01192443,
      "auxiliary_loss_mlp": 0.00902342,
      "balance_loss_clip": 1.05868578,
      "balance_loss_mlp": 1.00158966,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.811491320139932,
      "language_loss": 0.70152283,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.7224707,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.685079336166382
    },
    {
      "auxiliary_loss_clip": 0.01201141,
      "auxiliary_loss_mlp": 0.01032251,
      "balance_loss_clip": 1.05872488,
      "balance_loss_mlp": 1.02416873,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.623043495614768,
      "language_loss": 0.78055358,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80288756,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.637237787246704
    },
    {
      "auxiliary_loss_clip": 0.01199743,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.05704629,
      "balance_loss_mlp": 1.02030218,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.2125436224281883,
      "language_loss": 0.86300743,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.88529557,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 2.5965752601623535
    },
    {
      "auxiliary_loss_clip": 0.01173672,
      "auxiliary_loss_mlp": 0.01030888,
      "balance_loss_clip": 1.05522704,
      "balance_loss_mlp": 1.0215714,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.8422487780773131,
      "language_loss": 0.80607367,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82811928,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.621411085128784
    },
    {
      "auxiliary_loss_clip": 0.01198466,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.05778146,
      "balance_loss_mlp": 1.02194142,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 4.414640199656265,
      "language_loss": 0.86174798,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88403457,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.5944314002990723
    },
    {
      "auxiliary_loss_clip": 0.01188199,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.05627561,
      "balance_loss_mlp": 1.02200794,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.8644402812980496,
      "language_loss": 0.71257746,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73476303,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.5973992347717285
    },
    {
      "auxiliary_loss_clip": 0.01159259,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.05333638,
      "balance_loss_mlp": 1.02161121,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 2.0721451309954966,
      "language_loss": 0.81026626,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83217126,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.7135684490203857
    },
    {
      "auxiliary_loss_clip": 0.01189745,
      "auxiliary_loss_mlp": 0.00902567,
      "balance_loss_clip": 1.05626082,
      "balance_loss_mlp": 1.00172293,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.2530018217019525,
      "language_loss": 0.75456941,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77549255,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.6172902584075928
    },
    {
      "auxiliary_loss_clip": 0.0120185,
      "auxiliary_loss_mlp": 0.01037752,
      "balance_loss_clip": 1.05856204,
      "balance_loss_mlp": 1.02792287,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.8932386824901595,
      "language_loss": 0.82162493,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84402096,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.5907297134399414
    },
    {
      "auxiliary_loss_clip": 0.01198871,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.05507326,
      "balance_loss_mlp": 1.01872826,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 2.020167010667524,
      "language_loss": 0.7876513,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80991757,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.5937931537628174
    },
    {
      "auxiliary_loss_clip": 0.01174784,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.05455375,
      "balance_loss_mlp": 1.01998401,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.707014430240977,
      "language_loss": 0.78269374,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80472881,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.7453551292419434
    },
    {
      "auxiliary_loss_clip": 0.0115127,
      "auxiliary_loss_mlp": 0.01036927,
      "balance_loss_clip": 1.05040872,
      "balance_loss_mlp": 1.02778959,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.7495446115546422,
      "language_loss": 0.80135834,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82324028,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.7122864723205566
    },
    {
      "auxiliary_loss_clip": 0.01088873,
      "auxiliary_loss_mlp": 0.0100493,
      "balance_loss_clip": 1.02404189,
      "balance_loss_mlp": 1.00341058,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7826126193403092,
      "language_loss": 0.61617064,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63710868,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.296342372894287
    },
    {
      "auxiliary_loss_clip": 0.01206827,
      "auxiliary_loss_mlp": 0.0103529,
      "balance_loss_clip": 1.06182396,
      "balance_loss_mlp": 1.02609348,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 2.2199585452118313,
      "language_loss": 0.76750994,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78993106,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 3.6729586124420166
    },
    {
      "auxiliary_loss_clip": 0.01185609,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.05357778,
      "balance_loss_mlp": 1.02430868,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.961174305538444,
      "language_loss": 0.84450471,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.86669046,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.607274293899536
    },
    {
      "auxiliary_loss_clip": 0.01192115,
      "auxiliary_loss_mlp": 0.01036553,
      "balance_loss_clip": 1.05894887,
      "balance_loss_mlp": 1.02711153,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.8878881951616542,
      "language_loss": 0.72982597,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.75211263,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.6716394424438477
    },
    {
      "auxiliary_loss_clip": 0.01161447,
      "auxiliary_loss_mlp": 0.01036046,
      "balance_loss_clip": 1.04865181,
      "balance_loss_mlp": 1.02636075,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 2.342593546752342,
      "language_loss": 0.78116572,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.8031407,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 4.478054761886597
    },
    {
      "auxiliary_loss_clip": 0.01177792,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.05675352,
      "balance_loss_mlp": 1.02115846,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.1753997821171582,
      "language_loss": 0.73392987,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.75601339,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.7411584854125977
    },
    {
      "auxiliary_loss_clip": 0.01093132,
      "auxiliary_loss_mlp": 0.00891484,
      "balance_loss_clip": 1.02616143,
      "balance_loss_mlp": 1.0002445,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8310500861193431,
      "language_loss": 0.62392175,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64376789,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.179309129714966
    },
    {
      "auxiliary_loss_clip": 0.01165248,
      "auxiliary_loss_mlp": 0.01033855,
      "balance_loss_clip": 1.05167234,
      "balance_loss_mlp": 1.02468181,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 2.204409574491108,
      "language_loss": 0.83639348,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.85838449,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.76523494720459
    },
    {
      "auxiliary_loss_clip": 0.01196129,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 1.05936766,
      "balance_loss_mlp": 1.01673925,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 1.869653500785101,
      "language_loss": 0.94539398,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.96761596,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.608124256134033
    },
    {
      "auxiliary_loss_clip": 0.01183299,
      "auxiliary_loss_mlp": 0.01033954,
      "balance_loss_clip": 1.05821538,
      "balance_loss_mlp": 1.02405369,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 2.2910925256199883,
      "language_loss": 0.85289383,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.8750664,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 3.588641405105591
    },
    {
      "auxiliary_loss_clip": 0.01203393,
      "auxiliary_loss_mlp": 0.01039874,
      "balance_loss_clip": 1.06007421,
      "balance_loss_mlp": 1.02974164,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.568296656148826,
      "language_loss": 0.83954322,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.86197585,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.4801249504089355
    },
    {
      "auxiliary_loss_clip": 0.01186434,
      "auxiliary_loss_mlp": 0.01030722,
      "balance_loss_clip": 1.05485749,
      "balance_loss_mlp": 1.02167988,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 2.0874755772812503,
      "language_loss": 0.68477923,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70695078,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.6396398544311523
    },
    {
      "auxiliary_loss_clip": 0.0115527,
      "auxiliary_loss_mlp": 0.01035814,
      "balance_loss_clip": 1.05260599,
      "balance_loss_mlp": 1.0267365,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 14.492310988453191,
      "language_loss": 0.62164974,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.64356053,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.5976016521453857
    },
    {
      "auxiliary_loss_clip": 0.0118399,
      "auxiliary_loss_mlp": 0.01029038,
      "balance_loss_clip": 1.05726492,
      "balance_loss_mlp": 1.02080023,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.5968801633700929,
      "language_loss": 0.83745772,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85958797,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.6075544357299805
    },
    {
      "auxiliary_loss_clip": 0.01080463,
      "auxiliary_loss_mlp": 0.01001657,
      "balance_loss_clip": 1.02182913,
      "balance_loss_mlp": 1.0000838,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7760225868743834,
      "language_loss": 0.59474981,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61557102,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.0664355754852295
    },
    {
      "auxiliary_loss_clip": 0.01183701,
      "auxiliary_loss_mlp": 0.0103111,
      "balance_loss_clip": 1.0570389,
      "balance_loss_mlp": 1.02211523,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 2.5068675724847096,
      "language_loss": 0.66872936,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.6908775,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.6988425254821777
    },
    {
      "auxiliary_loss_clip": 0.01190864,
      "auxiliary_loss_mlp": 0.0103412,
      "balance_loss_clip": 1.05965877,
      "balance_loss_mlp": 1.02498865,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.9175940705058216,
      "language_loss": 0.82395995,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84620982,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.6904797554016113
    },
    {
      "auxiliary_loss_clip": 0.01185371,
      "auxiliary_loss_mlp": 0.01030149,
      "balance_loss_clip": 1.0541594,
      "balance_loss_mlp": 1.02056432,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.7512019375255508,
      "language_loss": 0.85863918,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.88079441,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.6569182872772217
    },
    {
      "auxiliary_loss_clip": 0.01191679,
      "auxiliary_loss_mlp": 0.01031611,
      "balance_loss_clip": 1.0579195,
      "balance_loss_mlp": 1.02289104,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.1610689468028057,
      "language_loss": 0.63051093,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65274382,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.7774572372436523
    },
    {
      "auxiliary_loss_clip": 0.0117247,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 1.05193567,
      "balance_loss_mlp": 1.02481735,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.7556435029331097,
      "language_loss": 0.81786335,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83991951,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.6794145107269287
    },
    {
      "auxiliary_loss_clip": 0.01152579,
      "auxiliary_loss_mlp": 0.01035072,
      "balance_loss_clip": 1.04778123,
      "balance_loss_mlp": 1.02623868,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 3.116954495855646,
      "language_loss": 0.86728954,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88916612,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.692173957824707
    },
    {
      "auxiliary_loss_clip": 0.01171558,
      "auxiliary_loss_mlp": 0.01039165,
      "balance_loss_clip": 1.05447745,
      "balance_loss_mlp": 1.03019428,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 2.8743256717283674,
      "language_loss": 0.7144655,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73657274,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.630462884902954
    },
    {
      "auxiliary_loss_clip": 0.01171285,
      "auxiliary_loss_mlp": 0.01035268,
      "balance_loss_clip": 1.05368829,
      "balance_loss_mlp": 1.02604079,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 1.910226787720981,
      "language_loss": 0.81165552,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.83372104,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.745056629180908
    },
    {
      "auxiliary_loss_clip": 0.01177847,
      "auxiliary_loss_mlp": 0.01026052,
      "balance_loss_clip": 1.05530155,
      "balance_loss_mlp": 1.01756477,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.717612536419315,
      "language_loss": 0.68410736,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70614642,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.678840160369873
    },
    {
      "auxiliary_loss_clip": 0.01201311,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 1.05812502,
      "balance_loss_mlp": 1.02261901,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 3.400025938835834,
      "language_loss": 0.72854376,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.75087678,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 2.7278852462768555
    },
    {
      "auxiliary_loss_clip": 0.01178755,
      "auxiliary_loss_mlp": 0.01030379,
      "balance_loss_clip": 1.05588627,
      "balance_loss_mlp": 1.02109814,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.2449465647328504,
      "language_loss": 0.87712109,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89921236,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.6723101139068604
    },
    {
      "auxiliary_loss_clip": 0.01201795,
      "auxiliary_loss_mlp": 0.01036722,
      "balance_loss_clip": 1.05715394,
      "balance_loss_mlp": 1.02706635,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 3.1519812195510872,
      "language_loss": 0.7080127,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.73039794,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.609236717224121
    },
    {
      "auxiliary_loss_clip": 0.01171046,
      "auxiliary_loss_mlp": 0.01031329,
      "balance_loss_clip": 1.05362463,
      "balance_loss_mlp": 1.02199471,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 2.2224665732669013,
      "language_loss": 0.86825693,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.89028072,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 3.6391522884368896
    },
    {
      "auxiliary_loss_clip": 0.01150595,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.04781973,
      "balance_loss_mlp": 1.02185643,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 2.173048137376856,
      "language_loss": 0.7991845,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.82099593,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.7159202098846436
    },
    {
      "auxiliary_loss_clip": 0.01201995,
      "auxiliary_loss_mlp": 0.01029625,
      "balance_loss_clip": 1.05849206,
      "balance_loss_mlp": 1.02069592,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 1.747343123214391,
      "language_loss": 0.8565163,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87883252,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.6438961029052734
    },
    {
      "auxiliary_loss_clip": 0.0118087,
      "auxiliary_loss_mlp": 0.01035024,
      "balance_loss_clip": 1.05631578,
      "balance_loss_mlp": 1.02608299,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.9823979794061142,
      "language_loss": 0.77781045,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79996943,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.707732677459717
    },
    {
      "auxiliary_loss_clip": 0.01176061,
      "auxiliary_loss_mlp": 0.01034643,
      "balance_loss_clip": 1.05454981,
      "balance_loss_mlp": 1.02498698,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 2.09510804836578,
      "language_loss": 0.73528081,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75738788,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 4.422220706939697
    },
    {
      "auxiliary_loss_clip": 0.01202127,
      "auxiliary_loss_mlp": 0.0103724,
      "balance_loss_clip": 1.05916572,
      "balance_loss_mlp": 1.02779317,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.422531629185507,
      "language_loss": 0.65258199,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.67497563,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.7344539165496826
    },
    {
      "auxiliary_loss_clip": 0.01183996,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.0564971,
      "balance_loss_mlp": 1.0183928,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 2.2579495173460904,
      "language_loss": 0.72153401,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74364412,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.648873805999756
    },
    {
      "auxiliary_loss_clip": 0.01197743,
      "auxiliary_loss_mlp": 0.01031823,
      "balance_loss_clip": 1.05709279,
      "balance_loss_mlp": 1.02294827,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.8248501023990502,
      "language_loss": 0.72747797,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.74977362,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.675081491470337
    },
    {
      "auxiliary_loss_clip": 0.01173481,
      "auxiliary_loss_mlp": 0.0090247,
      "balance_loss_clip": 1.05349064,
      "balance_loss_mlp": 1.00131226,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.143208371805735,
      "language_loss": 0.84555984,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.8663193,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.683999538421631
    },
    {
      "auxiliary_loss_clip": 0.01197203,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.05788565,
      "balance_loss_mlp": 1.02089238,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 2.124578271558873,
      "language_loss": 0.69820303,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.72047079,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 3.548128366470337
    },
    {
      "auxiliary_loss_clip": 0.01185288,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 1.05731678,
      "balance_loss_mlp": 1.02383852,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.720112713814062,
      "language_loss": 0.84482729,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86700737,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.663764238357544
    },
    {
      "auxiliary_loss_clip": 0.01191178,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.05926037,
      "balance_loss_mlp": 1.02671313,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 2.0608568962750624,
      "language_loss": 0.68190789,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70417392,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.5787339210510254
    },
    {
      "auxiliary_loss_clip": 0.0118405,
      "auxiliary_loss_mlp": 0.01038731,
      "balance_loss_clip": 1.05855727,
      "balance_loss_mlp": 1.02962971,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.675082445937793,
      "language_loss": 0.79583669,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81806445,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.819796085357666
    },
    {
      "auxiliary_loss_clip": 0.01180939,
      "auxiliary_loss_mlp": 0.01039286,
      "balance_loss_clip": 1.05362487,
      "balance_loss_mlp": 1.02939749,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 1.9892587489913263,
      "language_loss": 0.69714332,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.71934557,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.752572536468506
    },
    {
      "auxiliary_loss_clip": 0.01153482,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.0543468,
      "balance_loss_mlp": 1.02466083,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.1937978668454474,
      "language_loss": 0.89492691,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91679311,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.8169913291931152
    },
    {
      "auxiliary_loss_clip": 0.01176198,
      "auxiliary_loss_mlp": 0.01031674,
      "balance_loss_clip": 1.05503392,
      "balance_loss_mlp": 1.0228405,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 2.284499802503136,
      "language_loss": 0.8026889,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82476759,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.7137928009033203
    },
    {
      "auxiliary_loss_clip": 0.01155329,
      "auxiliary_loss_mlp": 0.01032103,
      "balance_loss_clip": 1.04995108,
      "balance_loss_mlp": 1.02293587,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 2.14499613120052,
      "language_loss": 0.8529259,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87480026,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.682034969329834
    },
    {
      "auxiliary_loss_clip": 0.01188879,
      "auxiliary_loss_mlp": 0.01037812,
      "balance_loss_clip": 1.05945551,
      "balance_loss_mlp": 1.02869201,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.8580079222607595,
      "language_loss": 0.84145534,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86372226,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.6249756813049316
    },
    {
      "auxiliary_loss_clip": 0.01154539,
      "auxiliary_loss_mlp": 0.01030609,
      "balance_loss_clip": 1.05282044,
      "balance_loss_mlp": 1.02193046,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.4143883425373436,
      "language_loss": 0.79390931,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81576085,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.7214581966400146
    },
    {
      "auxiliary_loss_clip": 0.01192461,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.05645037,
      "balance_loss_mlp": 1.02092659,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 1.9023916789414521,
      "language_loss": 0.77138948,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79361361,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.5636377334594727
    },
    {
      "auxiliary_loss_clip": 0.01181104,
      "auxiliary_loss_mlp": 0.01027804,
      "balance_loss_clip": 1.05739212,
      "balance_loss_mlp": 1.01874387,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.2802278083163916,
      "language_loss": 0.81484652,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83693564,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.6905078887939453
    },
    {
      "auxiliary_loss_clip": 0.01178722,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 1.05889678,
      "balance_loss_mlp": 1.02337193,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 2.3452353478359345,
      "language_loss": 0.67886829,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70097399,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.6876022815704346
    },
    {
      "auxiliary_loss_clip": 0.01191387,
      "auxiliary_loss_mlp": 0.01030704,
      "balance_loss_clip": 1.05368185,
      "balance_loss_mlp": 1.02125049,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.4642918805659924,
      "language_loss": 0.82448304,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84670401,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.619374990463257
    },
    {
      "auxiliary_loss_clip": 0.01195865,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.05878043,
      "balance_loss_mlp": 1.02024782,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.9612412320774593,
      "language_loss": 0.8853091,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90755695,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 2.6084952354431152
    },
    {
      "auxiliary_loss_clip": 0.01170749,
      "auxiliary_loss_mlp": 0.00902469,
      "balance_loss_clip": 1.05493236,
      "balance_loss_mlp": 1.00168777,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.8726954151750534,
      "language_loss": 0.86164916,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.88238132,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.7391133308410645
    },
    {
      "auxiliary_loss_clip": 0.01151988,
      "auxiliary_loss_mlp": 0.01029859,
      "balance_loss_clip": 1.04984713,
      "balance_loss_mlp": 1.02152586,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.383303384360769,
      "language_loss": 0.52353704,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54535556,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.697535276412964
    },
    {
      "auxiliary_loss_clip": 0.01196198,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.05665064,
      "balance_loss_mlp": 1.01747704,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 2.197293359203425,
      "language_loss": 0.83608121,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.85830092,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.598785161972046
    },
    {
      "auxiliary_loss_clip": 0.01189693,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.05548024,
      "balance_loss_mlp": 1.01859689,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 2.7411834538264745,
      "language_loss": 0.90739071,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.9295646,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 3.4942033290863037
    },
    {
      "auxiliary_loss_clip": 0.01173038,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 1.05386662,
      "balance_loss_mlp": 1.01784432,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.2829234027747534,
      "language_loss": 0.85654509,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.87853742,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.6211326122283936
    },
    {
      "auxiliary_loss_clip": 0.0119134,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.05545831,
      "balance_loss_mlp": 1.02321422,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 1.8149605266145274,
      "language_loss": 0.81388533,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83611619,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.726947546005249
    },
    {
      "auxiliary_loss_clip": 0.01173945,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.05392683,
      "balance_loss_mlp": 1.02157438,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.828942878447945,
      "language_loss": 0.79776555,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.8198095,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 3.6558332443237305
    },
    {
      "auxiliary_loss_clip": 0.01198591,
      "auxiliary_loss_mlp": 0.01030033,
      "balance_loss_clip": 1.06012392,
      "balance_loss_mlp": 1.02138448,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 2.0744398422662376,
      "language_loss": 0.76319408,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78548032,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 3.575117826461792
    },
    {
      "auxiliary_loss_clip": 0.01186716,
      "auxiliary_loss_mlp": 0.01034284,
      "balance_loss_clip": 1.05672407,
      "balance_loss_mlp": 1.02599335,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.5567980577266785,
      "language_loss": 0.73179293,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75400299,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.6568753719329834
    },
    {
      "auxiliary_loss_clip": 0.01177796,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.05749822,
      "balance_loss_mlp": 1.02133846,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 1.8967187391867548,
      "language_loss": 0.7158494,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73793608,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.830789089202881
    },
    {
      "auxiliary_loss_clip": 0.01172487,
      "auxiliary_loss_mlp": 0.01031579,
      "balance_loss_clip": 1.05681205,
      "balance_loss_mlp": 1.02300131,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.655884110601936,
      "language_loss": 0.85832983,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.8803705,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.7393805980682373
    },
    {
      "auxiliary_loss_clip": 0.01200548,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 1.0592587,
      "balance_loss_mlp": 1.01980901,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.9484831449492581,
      "language_loss": 0.80095512,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82325172,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 3.5360002517700195
    },
    {
      "auxiliary_loss_clip": 0.01197297,
      "auxiliary_loss_mlp": 0.00902679,
      "balance_loss_clip": 1.05808306,
      "balance_loss_mlp": 1.00157499,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.7843216363616692,
      "language_loss": 0.828789,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.84978878,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.5649538040161133
    },
    {
      "auxiliary_loss_clip": 0.01182454,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.05508256,
      "balance_loss_mlp": 1.01957178,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 15.041657899906344,
      "language_loss": 0.76835126,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.79047114,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.711153507232666
    },
    {
      "auxiliary_loss_clip": 0.01187135,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.05310392,
      "balance_loss_mlp": 1.01675797,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 1.9317295542452486,
      "language_loss": 0.87468433,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89681518,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.6185481548309326
    },
    {
      "auxiliary_loss_clip": 0.01199904,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.06059635,
      "balance_loss_mlp": 1.01782465,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 2.7561470392433804,
      "language_loss": 0.72807074,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.7503382,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.6165146827697754
    },
    {
      "auxiliary_loss_clip": 0.01183535,
      "auxiliary_loss_mlp": 0.01029074,
      "balance_loss_clip": 1.05790949,
      "balance_loss_mlp": 1.02030075,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 2.579783043722063,
      "language_loss": 0.85056603,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87269211,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.687230348587036
    },
    {
      "auxiliary_loss_clip": 0.01187623,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.0545435,
      "balance_loss_mlp": 1.0179739,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 3.42238795947055,
      "language_loss": 0.89889693,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.92103928,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 2.6657328605651855
    },
    {
      "auxiliary_loss_clip": 0.01165101,
      "auxiliary_loss_mlp": 0.00903468,
      "balance_loss_clip": 1.04896259,
      "balance_loss_mlp": 1.0016284,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.6833253303270395,
      "language_loss": 0.66361731,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68430305,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.7901711463928223
    },
    {
      "auxiliary_loss_clip": 0.01180035,
      "auxiliary_loss_mlp": 0.01032204,
      "balance_loss_clip": 1.05779123,
      "balance_loss_mlp": 1.0225153,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.7463116620244048,
      "language_loss": 0.77366543,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79578781,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.6642160415649414
    },
    {
      "auxiliary_loss_clip": 0.01198753,
      "auxiliary_loss_mlp": 0.01035299,
      "balance_loss_clip": 1.05827236,
      "balance_loss_mlp": 1.02632201,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 2.448675317353801,
      "language_loss": 0.74691784,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76925838,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.5616140365600586
    },
    {
      "auxiliary_loss_clip": 0.01158059,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.05395818,
      "balance_loss_mlp": 1.02198434,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 2.210135977605548,
      "language_loss": 0.72215068,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74403727,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.700533390045166
    },
    {
      "auxiliary_loss_clip": 0.01185714,
      "auxiliary_loss_mlp": 0.00902844,
      "balance_loss_clip": 1.05931067,
      "balance_loss_mlp": 1.00161099,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 1.9397588215968784,
      "language_loss": 0.88952374,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.91040933,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.6965413093566895
    },
    {
      "auxiliary_loss_clip": 0.01188153,
      "auxiliary_loss_mlp": 0.01030809,
      "balance_loss_clip": 1.0539093,
      "balance_loss_mlp": 1.02254748,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 2.391665733040084,
      "language_loss": 0.80411673,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82630634,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.645324468612671
    },
    {
      "auxiliary_loss_clip": 0.01175598,
      "auxiliary_loss_mlp": 0.0103328,
      "balance_loss_clip": 1.05725932,
      "balance_loss_mlp": 1.0238986,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 2.010512740792507,
      "language_loss": 0.78194624,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80403501,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.717327833175659
    },
    {
      "auxiliary_loss_clip": 0.01185897,
      "auxiliary_loss_mlp": 0.01028279,
      "balance_loss_clip": 1.0543685,
      "balance_loss_mlp": 1.01913595,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.083456377759079,
      "language_loss": 0.70350826,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72564995,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 2.6707310676574707
    },
    {
      "auxiliary_loss_clip": 0.0118449,
      "auxiliary_loss_mlp": 0.01030211,
      "balance_loss_clip": 1.05329657,
      "balance_loss_mlp": 1.02093041,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 1.9397199770177511,
      "language_loss": 0.92278481,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94493186,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.6479005813598633
    },
    {
      "auxiliary_loss_clip": 0.0119426,
      "auxiliary_loss_mlp": 0.010402,
      "balance_loss_clip": 1.05699182,
      "balance_loss_mlp": 1.03035343,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 3.1070126150697854,
      "language_loss": 0.77940249,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.80174708,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.5953996181488037
    },
    {
      "auxiliary_loss_clip": 0.01173377,
      "auxiliary_loss_mlp": 0.00902173,
      "balance_loss_clip": 1.05228162,
      "balance_loss_mlp": 1.00166166,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.9648229341524561,
      "language_loss": 0.64551961,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.66627508,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.694169759750366
    },
    {
      "auxiliary_loss_clip": 0.01191384,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.05808544,
      "balance_loss_mlp": 1.02033174,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.5396496843472685,
      "language_loss": 0.71887761,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.74109077,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 3.6059107780456543
    },
    {
      "auxiliary_loss_clip": 0.01194634,
      "auxiliary_loss_mlp": 0.01035122,
      "balance_loss_clip": 1.06263614,
      "balance_loss_mlp": 1.02569866,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 29.597768778867582,
      "language_loss": 0.67641914,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69871664,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.586294412612915
    },
    {
      "auxiliary_loss_clip": 0.01185034,
      "auxiliary_loss_mlp": 0.01034272,
      "balance_loss_clip": 1.05802202,
      "balance_loss_mlp": 1.02496731,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 1.784576268826782,
      "language_loss": 0.6513468,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67353988,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.868987560272217
    },
    {
      "auxiliary_loss_clip": 0.01170213,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.05109715,
      "balance_loss_mlp": 1.02211308,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 1.9316582092843833,
      "language_loss": 0.65208709,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67409682,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.7407386302948
    },
    {
      "auxiliary_loss_clip": 0.01201398,
      "auxiliary_loss_mlp": 0.01042533,
      "balance_loss_clip": 1.06142676,
      "balance_loss_mlp": 1.03305614,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.8448011854493152,
      "language_loss": 0.80066198,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82310128,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 3.5615651607513428
    },
    {
      "auxiliary_loss_clip": 0.01071201,
      "auxiliary_loss_mlp": 0.01003844,
      "balance_loss_clip": 1.02567506,
      "balance_loss_mlp": 1.00225818,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.8048202909735508,
      "language_loss": 0.6024121,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62316251,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 4.12193751335144
    },
    {
      "auxiliary_loss_clip": 0.01163958,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.05606413,
      "balance_loss_mlp": 1.02342868,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.7421557695016527,
      "language_loss": 0.76515532,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.78711152,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.6725783348083496
    },
    {
      "auxiliary_loss_clip": 0.01204682,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.06287587,
      "balance_loss_mlp": 1.02119005,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.6345733462902046,
      "language_loss": 0.7924552,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81480652,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.6410458087921143
    },
    {
      "auxiliary_loss_clip": 0.01174736,
      "auxiliary_loss_mlp": 0.01031392,
      "balance_loss_clip": 1.05444729,
      "balance_loss_mlp": 1.02211797,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.854849277786792,
      "language_loss": 0.75457656,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77663785,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.744565010070801
    },
    {
      "auxiliary_loss_clip": 0.01192319,
      "auxiliary_loss_mlp": 0.01028076,
      "balance_loss_clip": 1.05647874,
      "balance_loss_mlp": 1.01955879,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 2.064756930731109,
      "language_loss": 0.68705654,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.70926046,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 3.5950756072998047
    },
    {
      "auxiliary_loss_clip": 0.01138425,
      "auxiliary_loss_mlp": 0.01030857,
      "balance_loss_clip": 1.0501864,
      "balance_loss_mlp": 1.02124321,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 3.391868605563749,
      "language_loss": 0.66674781,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.68844068,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.84782338142395
    },
    {
      "auxiliary_loss_clip": 0.01175778,
      "auxiliary_loss_mlp": 0.01036771,
      "balance_loss_clip": 1.05976748,
      "balance_loss_mlp": 1.02770519,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.6784693799833703,
      "language_loss": 0.74642837,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76855379,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.8145532608032227
    },
    {
      "auxiliary_loss_clip": 0.01185711,
      "auxiliary_loss_mlp": 0.01030045,
      "balance_loss_clip": 1.05596137,
      "balance_loss_mlp": 1.02057362,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.7245296052047936,
      "language_loss": 0.73870862,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.76086617,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.716125249862671
    },
    {
      "auxiliary_loss_clip": 0.01183701,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.05888641,
      "balance_loss_mlp": 1.02250755,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 2.120682244119427,
      "language_loss": 0.74200022,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.76414984,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 2.677999496459961
    },
    {
      "auxiliary_loss_clip": 0.01205416,
      "auxiliary_loss_mlp": 0.01036563,
      "balance_loss_clip": 1.06276131,
      "balance_loss_mlp": 1.02728295,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.2734819529540133,
      "language_loss": 0.75398505,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.77640486,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.6746952533721924
    },
    {
      "auxiliary_loss_clip": 0.01194261,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.05708361,
      "balance_loss_mlp": 1.02197957,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 3.6241293752986232,
      "language_loss": 0.79084027,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81309688,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.6593728065490723
    },
    {
      "auxiliary_loss_clip": 0.0117931,
      "auxiliary_loss_mlp": 0.01033362,
      "balance_loss_clip": 1.0590713,
      "balance_loss_mlp": 1.02468944,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.6598503862407945,
      "language_loss": 0.71879208,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.74091876,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.708813428878784
    },
    {
      "auxiliary_loss_clip": 0.01176981,
      "auxiliary_loss_mlp": 0.01036948,
      "balance_loss_clip": 1.05589378,
      "balance_loss_mlp": 1.02757215,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.0441687676380274,
      "language_loss": 0.8183533,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84049261,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.6752190589904785
    },
    {
      "auxiliary_loss_clip": 0.01175859,
      "auxiliary_loss_mlp": 0.01034346,
      "balance_loss_clip": 1.0569545,
      "balance_loss_mlp": 1.02582264,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 3.035209134627907,
      "language_loss": 0.61663258,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.6387347,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.749847888946533
    },
    {
      "auxiliary_loss_clip": 0.01178127,
      "auxiliary_loss_mlp": 0.00903266,
      "balance_loss_clip": 1.05507219,
      "balance_loss_mlp": 1.00194621,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.6462154831715328,
      "language_loss": 0.74476707,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.76558101,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.682020425796509
    },
    {
      "auxiliary_loss_clip": 0.01177843,
      "auxiliary_loss_mlp": 0.01033465,
      "balance_loss_clip": 1.05770075,
      "balance_loss_mlp": 1.02438736,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 5.125552227663413,
      "language_loss": 0.73116744,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75328052,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.697299003601074
    },
    {
      "auxiliary_loss_clip": 0.01173572,
      "auxiliary_loss_mlp": 0.01034381,
      "balance_loss_clip": 1.05389142,
      "balance_loss_mlp": 1.02562475,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.48559250360152,
      "language_loss": 0.70510131,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.72718084,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.727860450744629
    },
    {
      "auxiliary_loss_clip": 0.01181076,
      "auxiliary_loss_mlp": 0.00903678,
      "balance_loss_clip": 1.06030655,
      "balance_loss_mlp": 1.00189698,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.431233874087697,
      "language_loss": 0.65426111,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.67510867,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.7095768451690674
    },
    {
      "auxiliary_loss_clip": 0.01183949,
      "auxiliary_loss_mlp": 0.01031422,
      "balance_loss_clip": 1.05992019,
      "balance_loss_mlp": 1.0224632,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 1.8915157390085502,
      "language_loss": 0.8056159,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82776964,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.7355828285217285
    },
    {
      "auxiliary_loss_clip": 0.01184327,
      "auxiliary_loss_mlp": 0.01032023,
      "balance_loss_clip": 1.05835629,
      "balance_loss_mlp": 1.02277863,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.80750455206488,
      "language_loss": 0.74697685,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.76914036,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.5880932807922363
    },
    {
      "auxiliary_loss_clip": 0.01188204,
      "auxiliary_loss_mlp": 0.0103435,
      "balance_loss_clip": 1.05766153,
      "balance_loss_mlp": 1.02449763,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.6890247090934927,
      "language_loss": 0.78912187,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81134742,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.717496633529663
    },
    {
      "auxiliary_loss_clip": 0.01196142,
      "auxiliary_loss_mlp": 0.01031662,
      "balance_loss_clip": 1.06619811,
      "balance_loss_mlp": 1.02263796,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.7921601428695977,
      "language_loss": 0.77926826,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80154634,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 3.489769458770752
    },
    {
      "auxiliary_loss_clip": 0.01193969,
      "auxiliary_loss_mlp": 0.01029265,
      "balance_loss_clip": 1.06464088,
      "balance_loss_mlp": 1.01889396,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.6341915419362607,
      "language_loss": 0.70747262,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72970498,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.689603328704834
    },
    {
      "auxiliary_loss_clip": 0.01174665,
      "auxiliary_loss_mlp": 0.0103909,
      "balance_loss_clip": 1.05374575,
      "balance_loss_mlp": 1.02873719,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 4.27733552838462,
      "language_loss": 0.70461857,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72675622,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.6265900135040283
    },
    {
      "auxiliary_loss_clip": 0.01196524,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.05833077,
      "balance_loss_mlp": 1.02430832,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.8492565575965134,
      "language_loss": 0.7794928,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80179358,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.6515815258026123
    },
    {
      "auxiliary_loss_clip": 0.01182419,
      "auxiliary_loss_mlp": 0.01030067,
      "balance_loss_clip": 1.05701685,
      "balance_loss_mlp": 1.02075076,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.8636376327247803,
      "language_loss": 0.84608889,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86821365,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 3.6216092109680176
    },
    {
      "auxiliary_loss_clip": 0.01164118,
      "auxiliary_loss_mlp": 0.01034661,
      "balance_loss_clip": 1.05037808,
      "balance_loss_mlp": 1.02557087,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.4927618968369316,
      "language_loss": 0.75750339,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.77949107,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 3.5818583965301514
    },
    {
      "auxiliary_loss_clip": 0.01154601,
      "auxiliary_loss_mlp": 0.01038903,
      "balance_loss_clip": 1.05101895,
      "balance_loss_mlp": 1.02992105,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 3.25457871971149,
      "language_loss": 0.89414716,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91608226,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.8232717514038086
    },
    {
      "auxiliary_loss_clip": 0.01173641,
      "auxiliary_loss_mlp": 0.01030317,
      "balance_loss_clip": 1.05697012,
      "balance_loss_mlp": 1.02129257,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.3496256166393055,
      "language_loss": 0.67460883,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.69664836,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.665943145751953
    },
    {
      "auxiliary_loss_clip": 0.01200081,
      "auxiliary_loss_mlp": 0.01035476,
      "balance_loss_clip": 1.05744553,
      "balance_loss_mlp": 1.02595091,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 1.8456264904757504,
      "language_loss": 0.80686736,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82922292,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.599982976913452
    },
    {
      "auxiliary_loss_clip": 0.01192479,
      "auxiliary_loss_mlp": 0.01032124,
      "balance_loss_clip": 1.06132793,
      "balance_loss_mlp": 1.02364826,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.0541719726037173,
      "language_loss": 0.76874119,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79098725,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 3.5551917552948
    },
    {
      "auxiliary_loss_clip": 0.01134023,
      "auxiliary_loss_mlp": 0.01036691,
      "balance_loss_clip": 1.04753804,
      "balance_loss_mlp": 1.0268209,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.582703768482186,
      "language_loss": 0.77153742,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.7932446,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.7745275497436523
    },
    {
      "auxiliary_loss_clip": 0.01173084,
      "auxiliary_loss_mlp": 0.01037687,
      "balance_loss_clip": 1.06204796,
      "balance_loss_mlp": 1.02889562,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 4.213293691508197,
      "language_loss": 0.8393566,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86146426,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.6787281036376953
    },
    {
      "auxiliary_loss_clip": 0.01174594,
      "auxiliary_loss_mlp": 0.01030204,
      "balance_loss_clip": 1.05441165,
      "balance_loss_mlp": 1.02095973,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.599428164290494,
      "language_loss": 0.61824208,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64029002,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.7959134578704834
    },
    {
      "auxiliary_loss_clip": 0.0117794,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.05544853,
      "balance_loss_mlp": 1.02357626,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 5.9910078973135,
      "language_loss": 0.69968337,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72179407,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.724604368209839
    },
    {
      "auxiliary_loss_clip": 0.01195563,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.05680895,
      "balance_loss_mlp": 1.0216558,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 2.922232912388629,
      "language_loss": 0.83282828,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85510099,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.6873786449432373
    },
    {
      "auxiliary_loss_clip": 0.01179113,
      "auxiliary_loss_mlp": 0.01031563,
      "balance_loss_clip": 1.0575701,
      "balance_loss_mlp": 1.02312911,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.7120297439487133,
      "language_loss": 0.7390129,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.76111972,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.748389482498169
    },
    {
      "auxiliary_loss_clip": 0.01184456,
      "auxiliary_loss_mlp": 0.01033542,
      "balance_loss_clip": 1.05626667,
      "balance_loss_mlp": 1.0246253,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 1.9383479638484824,
      "language_loss": 0.77653551,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79871547,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.727191209793091
    },
    {
      "auxiliary_loss_clip": 0.01188335,
      "auxiliary_loss_mlp": 0.01033397,
      "balance_loss_clip": 1.0577842,
      "balance_loss_mlp": 1.02438474,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 1.9203492063457497,
      "language_loss": 0.87917465,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90139198,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.6488375663757324
    },
    {
      "auxiliary_loss_clip": 0.01151515,
      "auxiliary_loss_mlp": 0.01031487,
      "balance_loss_clip": 1.0524255,
      "balance_loss_mlp": 1.02171159,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.15941941404419,
      "language_loss": 0.60483527,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62666529,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.8796277046203613
    },
    {
      "auxiliary_loss_clip": 0.01166897,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.05120564,
      "balance_loss_mlp": 1.01993918,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 4.29492504203543,
      "language_loss": 0.77207708,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.79403651,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.753933906555176
    },
    {
      "auxiliary_loss_clip": 0.01151629,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.0464375,
      "balance_loss_mlp": 1.02321696,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.0505337984073417,
      "language_loss": 0.66840005,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69024026,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.804645299911499
    },
    {
      "auxiliary_loss_clip": 0.01171364,
      "auxiliary_loss_mlp": 0.00903023,
      "balance_loss_clip": 1.05039597,
      "balance_loss_mlp": 1.00191116,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.2004459484322614,
      "language_loss": 0.84082294,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.86156678,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.686565637588501
    },
    {
      "auxiliary_loss_clip": 0.01190637,
      "auxiliary_loss_mlp": 0.01029023,
      "balance_loss_clip": 1.06006408,
      "balance_loss_mlp": 1.02050543,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 3.6766067773564073,
      "language_loss": 0.82922482,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.85142148,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 2.6130917072296143
    },
    {
      "auxiliary_loss_clip": 0.01187191,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.05623376,
      "balance_loss_mlp": 1.01774967,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 1.8546864686637667,
      "language_loss": 0.78817439,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.81030798,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.658916473388672
    },
    {
      "auxiliary_loss_clip": 0.01166952,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.05337512,
      "balance_loss_mlp": 1.01809216,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.516274204100419,
      "language_loss": 0.86806786,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.89000261,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.7087557315826416
    },
    {
      "auxiliary_loss_clip": 0.01169128,
      "auxiliary_loss_mlp": 0.0104105,
      "balance_loss_clip": 1.05558515,
      "balance_loss_mlp": 1.03204322,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 2.1689445812923833,
      "language_loss": 0.67128718,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69338894,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.681942939758301
    },
    {
      "auxiliary_loss_clip": 0.01189415,
      "auxiliary_loss_mlp": 0.01034956,
      "balance_loss_clip": 1.05879116,
      "balance_loss_mlp": 1.02591443,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 2.902485446120528,
      "language_loss": 0.71286023,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73510396,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 3.613664388656616
    },
    {
      "auxiliary_loss_clip": 0.01185872,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.05505574,
      "balance_loss_mlp": 1.01947248,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 4.308105684083286,
      "language_loss": 0.76365238,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78579831,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.611300468444824
    },
    {
      "auxiliary_loss_clip": 0.01202598,
      "auxiliary_loss_mlp": 0.01034593,
      "balance_loss_clip": 1.06079018,
      "balance_loss_mlp": 1.02487111,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.7459776062154164,
      "language_loss": 0.82188559,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84425747,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.606795072555542
    },
    {
      "auxiliary_loss_clip": 0.01189094,
      "auxiliary_loss_mlp": 0.01032152,
      "balance_loss_clip": 1.0568943,
      "balance_loss_mlp": 1.0221622,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.360190598138326,
      "language_loss": 0.85626602,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87847841,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.6918704509735107
    },
    {
      "auxiliary_loss_clip": 0.01181663,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.06058192,
      "balance_loss_mlp": 1.02537298,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 1.9731828761032024,
      "language_loss": 0.79602659,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81818819,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.6017982959747314
    },
    {
      "auxiliary_loss_clip": 0.01182926,
      "auxiliary_loss_mlp": 0.01026424,
      "balance_loss_clip": 1.05513072,
      "balance_loss_mlp": 1.01751351,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 3.367227899781104,
      "language_loss": 0.80485588,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82694936,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 4.5483434200286865
    },
    {
      "auxiliary_loss_clip": 0.01178259,
      "auxiliary_loss_mlp": 0.01034977,
      "balance_loss_clip": 1.05494046,
      "balance_loss_mlp": 1.02585137,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 2.4969308913709707,
      "language_loss": 0.72589076,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74802303,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.652078628540039
    },
    {
      "auxiliary_loss_clip": 0.0115408,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.05007541,
      "balance_loss_mlp": 1.02060509,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 1.998993891540672,
      "language_loss": 0.71892679,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.7407608,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.7578110694885254
    },
    {
      "auxiliary_loss_clip": 0.01170069,
      "auxiliary_loss_mlp": 0.01029417,
      "balance_loss_clip": 1.05199516,
      "balance_loss_mlp": 1.01970768,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 1.631049974466042,
      "language_loss": 0.75897086,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.78096569,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.7903552055358887
    },
    {
      "auxiliary_loss_clip": 0.01192082,
      "auxiliary_loss_mlp": 0.01032984,
      "balance_loss_clip": 1.05726445,
      "balance_loss_mlp": 1.02403164,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.06603975510962,
      "language_loss": 0.94472897,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.96697962,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.6149661540985107
    },
    {
      "auxiliary_loss_clip": 0.01184474,
      "auxiliary_loss_mlp": 0.01029571,
      "balance_loss_clip": 1.05888283,
      "balance_loss_mlp": 1.0203383,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 2.2784201245593496,
      "language_loss": 0.81485009,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83699054,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 3.5820162296295166
    },
    {
      "auxiliary_loss_clip": 0.01170618,
      "auxiliary_loss_mlp": 0.01033681,
      "balance_loss_clip": 1.05336261,
      "balance_loss_mlp": 1.02447534,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 1.893437655186272,
      "language_loss": 0.77121639,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79325938,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.719120502471924
    },
    {
      "auxiliary_loss_clip": 0.01186622,
      "auxiliary_loss_mlp": 0.01034256,
      "balance_loss_clip": 1.05665565,
      "balance_loss_mlp": 1.02518451,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.451120347817482,
      "language_loss": 0.80317712,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82538581,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.6378841400146484
    },
    {
      "auxiliary_loss_clip": 0.01179092,
      "auxiliary_loss_mlp": 0.01038052,
      "balance_loss_clip": 1.05756545,
      "balance_loss_mlp": 1.0287838,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.9109384951405506,
      "language_loss": 0.74668837,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76885974,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 2.7715444564819336
    },
    {
      "auxiliary_loss_clip": 0.01205476,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.06076837,
      "balance_loss_mlp": 1.0270505,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 1.9690336481343431,
      "language_loss": 0.71576226,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73819506,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.6335904598236084
    },
    {
      "auxiliary_loss_clip": 0.01193307,
      "auxiliary_loss_mlp": 0.01033104,
      "balance_loss_clip": 1.05979407,
      "balance_loss_mlp": 1.0237577,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.0923779644492995,
      "language_loss": 0.78570473,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80796891,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.6462607383728027
    },
    {
      "auxiliary_loss_clip": 0.01183305,
      "auxiliary_loss_mlp": 0.01038842,
      "balance_loss_clip": 1.0559957,
      "balance_loss_mlp": 1.02894187,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.8748424814755105,
      "language_loss": 0.80857909,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.83080053,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.636192560195923
    },
    {
      "auxiliary_loss_clip": 0.01185335,
      "auxiliary_loss_mlp": 0.0103514,
      "balance_loss_clip": 1.06011248,
      "balance_loss_mlp": 1.02501857,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.962667266240143,
      "language_loss": 0.81934297,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84154773,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.6691722869873047
    },
    {
      "auxiliary_loss_clip": 0.01193268,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.05754018,
      "balance_loss_mlp": 1.01906681,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 2.185331944060594,
      "language_loss": 0.72930294,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75151151,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.6038782596588135
    },
    {
      "auxiliary_loss_clip": 0.01175289,
      "auxiliary_loss_mlp": 0.00903728,
      "balance_loss_clip": 1.05578315,
      "balance_loss_mlp": 1.00188255,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.5944248700031003,
      "language_loss": 0.80647695,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82726717,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.7813730239868164
    },
    {
      "auxiliary_loss_clip": 0.01196804,
      "auxiliary_loss_mlp": 0.01030271,
      "balance_loss_clip": 1.06353712,
      "balance_loss_mlp": 1.02113926,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 2.197644812375207,
      "language_loss": 0.77565056,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.7979213,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.618579149246216
    },
    {
      "auxiliary_loss_clip": 0.01154245,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.04971087,
      "balance_loss_mlp": 1.01913297,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.3782310141102085,
      "language_loss": 0.81574273,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83757293,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.6665852069854736
    },
    {
      "auxiliary_loss_clip": 0.01193378,
      "auxiliary_loss_mlp": 0.01038888,
      "balance_loss_clip": 1.05702972,
      "balance_loss_mlp": 1.02874303,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 1.899063992385032,
      "language_loss": 0.77736634,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79968899,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 2.650512218475342
    },
    {
      "auxiliary_loss_clip": 0.01200425,
      "auxiliary_loss_mlp": 0.01033896,
      "balance_loss_clip": 1.06123149,
      "balance_loss_mlp": 1.02294087,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.517123165742279,
      "language_loss": 0.69370484,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71604806,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.657106637954712
    },
    {
      "auxiliary_loss_clip": 0.01189968,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.06023216,
      "balance_loss_mlp": 1.02249479,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 2.156321660531355,
      "language_loss": 0.69806951,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.72029173,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.6356616020202637
    },
    {
      "auxiliary_loss_clip": 0.01100341,
      "auxiliary_loss_mlp": 0.01010643,
      "balance_loss_clip": 1.02615523,
      "balance_loss_mlp": 1.00890243,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0660531675333613,
      "language_loss": 0.64750552,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66861534,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.1227951049804688
    },
    {
      "auxiliary_loss_clip": 0.01178865,
      "auxiliary_loss_mlp": 0.01031444,
      "balance_loss_clip": 1.05327773,
      "balance_loss_mlp": 1.02298045,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.5870358144042649,
      "language_loss": 0.92953295,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95163608,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.7435615062713623
    },
    {
      "auxiliary_loss_clip": 0.01191442,
      "auxiliary_loss_mlp": 0.01031786,
      "balance_loss_clip": 1.06013393,
      "balance_loss_mlp": 1.0220108,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.7804680637923211,
      "language_loss": 0.69581258,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71804488,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 3.57829213142395
    },
    {
      "auxiliary_loss_clip": 0.01167353,
      "auxiliary_loss_mlp": 0.01037026,
      "balance_loss_clip": 1.05356812,
      "balance_loss_mlp": 1.02753711,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 2.240124099536336,
      "language_loss": 0.80685866,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82890242,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.7566354274749756
    },
    {
      "auxiliary_loss_clip": 0.01188551,
      "auxiliary_loss_mlp": 0.0103723,
      "balance_loss_clip": 1.05485952,
      "balance_loss_mlp": 1.02729976,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 1.9351363330679787,
      "language_loss": 0.80011892,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.82237673,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.6453957557678223
    },
    {
      "auxiliary_loss_clip": 0.01193355,
      "auxiliary_loss_mlp": 0.01035056,
      "balance_loss_clip": 1.05932724,
      "balance_loss_mlp": 1.024822,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.737090695410112,
      "language_loss": 0.71037483,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.73265898,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.671064615249634
    },
    {
      "auxiliary_loss_clip": 0.0118176,
      "auxiliary_loss_mlp": 0.00903515,
      "balance_loss_clip": 1.05680954,
      "balance_loss_mlp": 1.00186968,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 1.900230462213631,
      "language_loss": 0.82790148,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84875429,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 3.605822801589966
    },
    {
      "auxiliary_loss_clip": 0.01090841,
      "auxiliary_loss_mlp": 0.01002665,
      "balance_loss_clip": 1.02382505,
      "balance_loss_mlp": 1.00099611,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8218501526002725,
      "language_loss": 0.63699239,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65792745,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.2089056968688965
    },
    {
      "auxiliary_loss_clip": 0.01191835,
      "auxiliary_loss_mlp": 0.01031856,
      "balance_loss_clip": 1.06064177,
      "balance_loss_mlp": 1.02199185,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.118356709621221,
      "language_loss": 0.83788586,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.8601228,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.681347131729126
    },
    {
      "auxiliary_loss_clip": 0.01174081,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.05483747,
      "balance_loss_mlp": 1.02237344,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.8724916094373136,
      "language_loss": 0.82841408,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.85047138,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.7108840942382812
    },
    {
      "auxiliary_loss_clip": 0.01165212,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.0548681,
      "balance_loss_mlp": 1.02874279,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.2778796444081997,
      "language_loss": 0.78602707,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.80805629,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.7677392959594727
    },
    {
      "auxiliary_loss_clip": 0.01199261,
      "auxiliary_loss_mlp": 0.01031009,
      "balance_loss_clip": 1.05990386,
      "balance_loss_mlp": 1.02137136,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 2.180165316216351,
      "language_loss": 0.75563413,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77793682,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.746428966522217
    },
    {
      "auxiliary_loss_clip": 0.01182555,
      "auxiliary_loss_mlp": 0.0103208,
      "balance_loss_clip": 1.05798197,
      "balance_loss_mlp": 1.02306759,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.212955070076795,
      "language_loss": 0.75721341,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77935976,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 3.5676651000976562
    },
    {
      "auxiliary_loss_clip": 0.01160105,
      "auxiliary_loss_mlp": 0.01034608,
      "balance_loss_clip": 1.05224442,
      "balance_loss_mlp": 1.02507782,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 2.4040168738456544,
      "language_loss": 0.76562864,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78757572,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.7823872566223145
    },
    {
      "auxiliary_loss_clip": 0.01163556,
      "auxiliary_loss_mlp": 0.01029773,
      "balance_loss_clip": 1.05147207,
      "balance_loss_mlp": 1.02020621,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.837907697885886,
      "language_loss": 0.68494439,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70687759,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.696199417114258
    },
    {
      "auxiliary_loss_clip": 0.01199258,
      "auxiliary_loss_mlp": 0.01033392,
      "balance_loss_clip": 1.05926645,
      "balance_loss_mlp": 1.02380764,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.7071559959895888,
      "language_loss": 0.69568425,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71801078,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.698671579360962
    },
    {
      "auxiliary_loss_clip": 0.01196865,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.05823219,
      "balance_loss_mlp": 1.02312398,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 1.9520830287193838,
      "language_loss": 0.72750568,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.74980056,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.5600922107696533
    },
    {
      "auxiliary_loss_clip": 0.0116614,
      "auxiliary_loss_mlp": 0.01034757,
      "balance_loss_clip": 1.0530076,
      "balance_loss_mlp": 1.02600658,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.782563851838877,
      "language_loss": 0.88427877,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90628779,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.7095866203308105
    },
    {
      "auxiliary_loss_clip": 0.01168812,
      "auxiliary_loss_mlp": 0.01034872,
      "balance_loss_clip": 1.05173552,
      "balance_loss_mlp": 1.02556217,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.9633689494664366,
      "language_loss": 0.63053071,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65256757,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.6660468578338623
    },
    {
      "auxiliary_loss_clip": 0.01203305,
      "auxiliary_loss_mlp": 0.01036728,
      "balance_loss_clip": 1.06003189,
      "balance_loss_mlp": 1.02688098,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.5292215604787054,
      "language_loss": 0.72267377,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74507409,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.639803171157837
    },
    {
      "auxiliary_loss_clip": 0.01159829,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.04924464,
      "balance_loss_mlp": 1.02390504,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 1.787861501179702,
      "language_loss": 0.73877949,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76071239,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.7697794437408447
    },
    {
      "auxiliary_loss_clip": 0.01143887,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.04860306,
      "balance_loss_mlp": 1.02197278,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 5.730544788730755,
      "language_loss": 0.68134153,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70309269,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.745924949645996
    },
    {
      "auxiliary_loss_clip": 0.01166495,
      "auxiliary_loss_mlp": 0.01025852,
      "balance_loss_clip": 1.05319631,
      "balance_loss_mlp": 1.0168159,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 2.67082331045384,
      "language_loss": 0.69247484,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.71439832,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.750969171524048
    },
    {
      "auxiliary_loss_clip": 0.01200708,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.0623759,
      "balance_loss_mlp": 1.0213182,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.4915950502248716,
      "language_loss": 0.86910349,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89141738,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.5919294357299805
    },
    {
      "auxiliary_loss_clip": 0.01142714,
      "auxiliary_loss_mlp": 0.0090362,
      "balance_loss_clip": 1.04759407,
      "balance_loss_mlp": 1.00215435,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.66436916760668,
      "language_loss": 0.77180916,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.79227245,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.8608763217926025
    },
    {
      "auxiliary_loss_clip": 0.01192723,
      "auxiliary_loss_mlp": 0.01036686,
      "balance_loss_clip": 1.05591345,
      "balance_loss_mlp": 1.02716732,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 2.210443547886436,
      "language_loss": 0.65054607,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67284018,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.616396188735962
    },
    {
      "auxiliary_loss_clip": 0.0119953,
      "auxiliary_loss_mlp": 0.01034865,
      "balance_loss_clip": 1.05880356,
      "balance_loss_mlp": 1.02483308,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 1.9377878232771173,
      "language_loss": 0.7982018,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.82054579,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.614537239074707
    },
    {
      "auxiliary_loss_clip": 0.01182399,
      "auxiliary_loss_mlp": 0.01029989,
      "balance_loss_clip": 1.06035268,
      "balance_loss_mlp": 1.02121568,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.174983854903827,
      "language_loss": 0.85719371,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.87931752,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.703580856323242
    },
    {
      "auxiliary_loss_clip": 0.01190213,
      "auxiliary_loss_mlp": 0.01029014,
      "balance_loss_clip": 1.05854309,
      "balance_loss_mlp": 1.02032411,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 2.4043607865706877,
      "language_loss": 0.7887007,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81089294,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.6460797786712646
    },
    {
      "auxiliary_loss_clip": 0.01169961,
      "auxiliary_loss_mlp": 0.00903236,
      "balance_loss_clip": 1.05186319,
      "balance_loss_mlp": 1.00195026,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 2.5252203650822804,
      "language_loss": 0.82799035,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84872234,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 3.686997413635254
    },
    {
      "auxiliary_loss_clip": 0.01176796,
      "auxiliary_loss_mlp": 0.0103132,
      "balance_loss_clip": 1.05161071,
      "balance_loss_mlp": 1.02191472,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.6197651941594997,
      "language_loss": 0.8025015,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.8245827,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.620154857635498
    },
    {
      "auxiliary_loss_clip": 0.01190566,
      "auxiliary_loss_mlp": 0.01030034,
      "balance_loss_clip": 1.05888915,
      "balance_loss_mlp": 1.02084279,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 2.1212254823984584,
      "language_loss": 0.71799588,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.74020183,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.621743679046631
    },
    {
      "auxiliary_loss_clip": 0.01201563,
      "auxiliary_loss_mlp": 0.01033566,
      "balance_loss_clip": 1.06045139,
      "balance_loss_mlp": 1.0238564,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.0316582583585103,
      "language_loss": 0.77896714,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.80131847,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.610440731048584
    },
    {
      "auxiliary_loss_clip": 0.01181067,
      "auxiliary_loss_mlp": 0.01030828,
      "balance_loss_clip": 1.05623949,
      "balance_loss_mlp": 1.02172661,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 5.451589975060252,
      "language_loss": 0.74439031,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76650929,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 4.543608903884888
    },
    {
      "auxiliary_loss_clip": 0.01090491,
      "auxiliary_loss_mlp": 0.01001817,
      "balance_loss_clip": 1.02519608,
      "balance_loss_mlp": 1.00026727,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.3822876783824636,
      "language_loss": 0.64549768,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66642082,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.2559592723846436
    },
    {
      "auxiliary_loss_clip": 0.01173913,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.05212879,
      "balance_loss_mlp": 1.0151428,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.1540089610116326,
      "language_loss": 0.70050323,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72249198,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.6431515216827393
    },
    {
      "auxiliary_loss_clip": 0.01084704,
      "auxiliary_loss_mlp": 0.0100262,
      "balance_loss_clip": 1.02496648,
      "balance_loss_mlp": 1.0010227,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.8106795341758427,
      "language_loss": 0.5776931,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59856635,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.240018367767334
    },
    {
      "auxiliary_loss_clip": 0.0116482,
      "auxiliary_loss_mlp": 0.01035097,
      "balance_loss_clip": 1.05694938,
      "balance_loss_mlp": 1.02666879,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.4678899672380807,
      "language_loss": 0.74077111,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.7627703,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.6905505657196045
    },
    {
      "auxiliary_loss_clip": 0.01157243,
      "auxiliary_loss_mlp": 0.01033518,
      "balance_loss_clip": 1.05363441,
      "balance_loss_mlp": 1.02397585,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.783875555146624,
      "language_loss": 0.83810353,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.8600111,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 3.673367977142334
    },
    {
      "auxiliary_loss_clip": 0.01195367,
      "auxiliary_loss_mlp": 0.01035946,
      "balance_loss_clip": 1.05757666,
      "balance_loss_mlp": 1.02704096,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.020688532859139,
      "language_loss": 0.71845073,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.74076384,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.6383657455444336
    },
    {
      "auxiliary_loss_clip": 0.01181475,
      "auxiliary_loss_mlp": 0.01032563,
      "balance_loss_clip": 1.05877233,
      "balance_loss_mlp": 1.02377093,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.9205583664804284,
      "language_loss": 0.62513673,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64727712,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.646937608718872
    },
    {
      "auxiliary_loss_clip": 0.01178081,
      "auxiliary_loss_mlp": 0.01025613,
      "balance_loss_clip": 1.05564451,
      "balance_loss_mlp": 1.0167675,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 3.657210674124123,
      "language_loss": 0.82121813,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84325504,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.7177584171295166
    },
    {
      "auxiliary_loss_clip": 0.01188834,
      "auxiliary_loss_mlp": 0.01029637,
      "balance_loss_clip": 1.05865169,
      "balance_loss_mlp": 1.02122092,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 2.9175400708687893,
      "language_loss": 0.63672531,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.65890998,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.6074817180633545
    },
    {
      "auxiliary_loss_clip": 0.01179369,
      "auxiliary_loss_mlp": 0.01036428,
      "balance_loss_clip": 1.05644739,
      "balance_loss_mlp": 1.02766669,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.6271260149485898,
      "language_loss": 0.75301552,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77517343,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.6773898601531982
    },
    {
      "auxiliary_loss_clip": 0.01176943,
      "auxiliary_loss_mlp": 0.01032764,
      "balance_loss_clip": 1.055668,
      "balance_loss_mlp": 1.02324486,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.177737958179676,
      "language_loss": 0.80806255,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.83015966,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.6333298683166504
    },
    {
      "auxiliary_loss_clip": 0.01186455,
      "auxiliary_loss_mlp": 0.01035833,
      "balance_loss_clip": 1.05515575,
      "balance_loss_mlp": 1.02626646,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.647755010513407,
      "language_loss": 0.69389808,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.71612096,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.6458537578582764
    },
    {
      "auxiliary_loss_clip": 0.01180534,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.05371356,
      "balance_loss_mlp": 1.01715326,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 2.052080520689624,
      "language_loss": 0.64681631,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.66887867,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.6317148208618164
    },
    {
      "auxiliary_loss_clip": 0.01199757,
      "auxiliary_loss_mlp": 0.01028994,
      "balance_loss_clip": 1.05816078,
      "balance_loss_mlp": 1.02007723,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 3.3722824379786243,
      "language_loss": 0.81423199,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83651954,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 2.567718744277954
    },
    {
      "auxiliary_loss_clip": 0.01157983,
      "auxiliary_loss_mlp": 0.01028685,
      "balance_loss_clip": 1.05076432,
      "balance_loss_mlp": 1.01967883,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 3.5344061976578116,
      "language_loss": 0.79673594,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81860262,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.6850669384002686
    },
    {
      "auxiliary_loss_clip": 0.01191018,
      "auxiliary_loss_mlp": 0.00902575,
      "balance_loss_clip": 1.05861044,
      "balance_loss_mlp": 1.00162911,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.2790647874698493,
      "language_loss": 0.73093522,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75187111,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.659923791885376
    },
    {
      "auxiliary_loss_clip": 0.01192155,
      "auxiliary_loss_mlp": 0.01037636,
      "balance_loss_clip": 1.05660725,
      "balance_loss_mlp": 1.02818227,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 1.8635154806826857,
      "language_loss": 0.80393469,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82623255,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.776714563369751
    },
    {
      "auxiliary_loss_clip": 0.01188851,
      "auxiliary_loss_mlp": 0.01031762,
      "balance_loss_clip": 1.05927837,
      "balance_loss_mlp": 1.02327406,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.7438777847901235,
      "language_loss": 0.79113656,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.81334269,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.6316275596618652
    },
    {
      "auxiliary_loss_clip": 0.01167085,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.05172467,
      "balance_loss_mlp": 1.0201571,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.5186087100077545,
      "language_loss": 0.72706848,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74902558,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.771606206893921
    },
    {
      "auxiliary_loss_clip": 0.01184037,
      "auxiliary_loss_mlp": 0.0102773,
      "balance_loss_clip": 1.05638361,
      "balance_loss_mlp": 1.01828241,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.7216921577587647,
      "language_loss": 0.81925988,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.8413775,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.6601781845092773
    },
    {
      "auxiliary_loss_clip": 0.01192357,
      "auxiliary_loss_mlp": 0.00902532,
      "balance_loss_clip": 1.05541134,
      "balance_loss_mlp": 1.00164354,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.255869670065189,
      "language_loss": 0.80083585,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.82178473,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.671644687652588
    },
    {
      "auxiliary_loss_clip": 0.0107897,
      "auxiliary_loss_mlp": 0.01002298,
      "balance_loss_clip": 1.03085542,
      "balance_loss_mlp": 1.00073624,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.7647804070327076,
      "language_loss": 0.59211683,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61292952,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 4.208177804946899
    },
    {
      "auxiliary_loss_clip": 0.01180493,
      "auxiliary_loss_mlp": 0.01030333,
      "balance_loss_clip": 1.05705833,
      "balance_loss_mlp": 1.02121341,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.8864757265214842,
      "language_loss": 0.70724511,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72935343,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.724932909011841
    },
    {
      "auxiliary_loss_clip": 0.0116591,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.05082631,
      "balance_loss_mlp": 1.021433,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 2.018946334753894,
      "language_loss": 0.82616538,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.84812701,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.750746965408325
    },
    {
      "auxiliary_loss_clip": 0.0120094,
      "auxiliary_loss_mlp": 0.01032652,
      "balance_loss_clip": 1.06074047,
      "balance_loss_mlp": 1.02427149,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 3.1723977081018435,
      "language_loss": 0.72871327,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75104916,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.637701988220215
    },
    {
      "auxiliary_loss_clip": 0.01164547,
      "auxiliary_loss_mlp": 0.01032783,
      "balance_loss_clip": 1.04957318,
      "balance_loss_mlp": 1.02408075,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 1.7848906655305303,
      "language_loss": 0.76930809,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79128134,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 3.782041549682617
    },
    {
      "auxiliary_loss_clip": 0.01189067,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.05950332,
      "balance_loss_mlp": 1.01857829,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.3919800632682593,
      "language_loss": 0.79747301,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81963575,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 3.604363203048706
    },
    {
      "auxiliary_loss_clip": 0.01184773,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.05539966,
      "balance_loss_mlp": 1.01613939,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.338364022638968,
      "language_loss": 0.80828601,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.8303777,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.5857813358306885
    },
    {
      "auxiliary_loss_clip": 0.01189392,
      "auxiliary_loss_mlp": 0.01033669,
      "balance_loss_clip": 1.05752134,
      "balance_loss_mlp": 1.0253365,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 2.401831442005881,
      "language_loss": 0.76858401,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79081458,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.653639554977417
    },
    {
      "auxiliary_loss_clip": 0.01185535,
      "auxiliary_loss_mlp": 0.01032396,
      "balance_loss_clip": 1.0554111,
      "balance_loss_mlp": 1.02342546,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 3.933142813935913,
      "language_loss": 0.83277881,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85495818,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.7127459049224854
    },
    {
      "auxiliary_loss_clip": 0.01183165,
      "auxiliary_loss_mlp": 0.01033648,
      "balance_loss_clip": 1.05763054,
      "balance_loss_mlp": 1.0249939,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.6023163684481752,
      "language_loss": 0.88533366,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90750182,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.688555955886841
    },
    {
      "auxiliary_loss_clip": 0.01179605,
      "auxiliary_loss_mlp": 0.01028767,
      "balance_loss_clip": 1.05588627,
      "balance_loss_mlp": 1.02119136,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 3.1312503147110253,
      "language_loss": 0.89378214,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.9158659,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 3.69659686088562
    },
    {
      "auxiliary_loss_clip": 0.01173465,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.05208063,
      "balance_loss_mlp": 1.0205462,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 4.068939584752165,
      "language_loss": 0.81711817,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.839149,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.7129175662994385
    },
    {
      "auxiliary_loss_clip": 0.01189293,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 1.0569613,
      "balance_loss_mlp": 1.01642799,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 1.8424928007806558,
      "language_loss": 0.71323842,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73537773,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.8559048175811768
    },
    {
      "auxiliary_loss_clip": 0.01173432,
      "auxiliary_loss_mlp": 0.01032067,
      "balance_loss_clip": 1.05553591,
      "balance_loss_mlp": 1.02430105,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.619352579944726,
      "language_loss": 0.89752477,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.91957974,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.6274187564849854
    },
    {
      "auxiliary_loss_clip": 0.01070997,
      "auxiliary_loss_mlp": 0.01001966,
      "balance_loss_clip": 1.02115655,
      "balance_loss_mlp": 1.00054789,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.7042139003228443,
      "language_loss": 0.56645107,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58718073,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.4700355529785156
    },
    {
      "auxiliary_loss_clip": 0.0116392,
      "auxiliary_loss_mlp": 0.01031039,
      "balance_loss_clip": 1.05138624,
      "balance_loss_mlp": 1.02275395,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 3.5011141017530156,
      "language_loss": 0.6697768,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69172639,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.722137451171875
    },
    {
      "auxiliary_loss_clip": 0.01156234,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.05072451,
      "balance_loss_mlp": 1.02304912,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 6.145783388634759,
      "language_loss": 0.7564851,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.77835917,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.6923344135284424
    },
    {
      "auxiliary_loss_clip": 0.01200828,
      "auxiliary_loss_mlp": 0.00902776,
      "balance_loss_clip": 1.05921698,
      "balance_loss_mlp": 1.00172174,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.5405195747038456,
      "language_loss": 0.82096469,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.84200072,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.621964693069458
    },
    {
      "auxiliary_loss_clip": 0.01178713,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.05379152,
      "balance_loss_mlp": 1.01692486,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 2.161809585566154,
      "language_loss": 0.83180034,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85383737,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 2.583082675933838
    },
    {
      "auxiliary_loss_clip": 0.01168177,
      "auxiliary_loss_mlp": 0.01027747,
      "balance_loss_clip": 1.05343819,
      "balance_loss_mlp": 1.01952159,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.6615473948189492,
      "language_loss": 0.79224318,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81420243,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.719836473464966
    },
    {
      "auxiliary_loss_clip": 0.0119062,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.05776834,
      "balance_loss_mlp": 1.0223043,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 4.2297217664929825,
      "language_loss": 0.77483821,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79705286,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.653876304626465
    },
    {
      "auxiliary_loss_clip": 0.0119573,
      "auxiliary_loss_mlp": 0.00901971,
      "balance_loss_clip": 1.05818021,
      "balance_loss_mlp": 1.00160551,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.521172721016656,
      "language_loss": 0.73978072,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76075774,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.5933215618133545
    },
    {
      "auxiliary_loss_clip": 0.01181361,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.05596042,
      "balance_loss_mlp": 1.02253461,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 1.9574585582346204,
      "language_loss": 0.79682267,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.818946,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.7018556594848633
    },
    {
      "auxiliary_loss_clip": 0.01165988,
      "auxiliary_loss_mlp": 0.01030755,
      "balance_loss_clip": 1.0558356,
      "balance_loss_mlp": 1.02237427,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.682298801428008,
      "language_loss": 0.69021869,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.71218616,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6978089809417725
    },
    {
      "auxiliary_loss_clip": 0.01172676,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.05442226,
      "balance_loss_mlp": 1.02127934,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.1154083936585337,
      "language_loss": 0.79290098,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.8149277,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.6403117179870605
    },
    {
      "auxiliary_loss_clip": 0.01179231,
      "auxiliary_loss_mlp": 0.01037083,
      "balance_loss_clip": 1.05863833,
      "balance_loss_mlp": 1.02865505,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 2.1025089439140063,
      "language_loss": 0.79116571,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81332886,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.695005178451538
    },
    {
      "auxiliary_loss_clip": 0.01168437,
      "auxiliary_loss_mlp": 0.01029358,
      "balance_loss_clip": 1.05361509,
      "balance_loss_mlp": 1.02125192,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.027247942606542,
      "language_loss": 0.76495981,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.78693771,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 3.6720099449157715
    },
    {
      "auxiliary_loss_clip": 0.01183803,
      "auxiliary_loss_mlp": 0.0102729,
      "balance_loss_clip": 1.05258226,
      "balance_loss_mlp": 1.01905894,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 1.8390089113976877,
      "language_loss": 0.76257306,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.784684,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.588899612426758
    },
    {
      "auxiliary_loss_clip": 0.01170656,
      "auxiliary_loss_mlp": 0.01033829,
      "balance_loss_clip": 1.05229354,
      "balance_loss_mlp": 1.02497804,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.7524995637229526,
      "language_loss": 0.74213266,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76417756,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.677638292312622
    },
    {
      "auxiliary_loss_clip": 0.0109718,
      "auxiliary_loss_mlp": 0.01005359,
      "balance_loss_clip": 1.02494454,
      "balance_loss_mlp": 1.00380945,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7949160283444366,
      "language_loss": 0.57671827,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59774363,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 2.951059579849243
    },
    {
      "auxiliary_loss_clip": 0.01163657,
      "auxiliary_loss_mlp": 0.0102898,
      "balance_loss_clip": 1.05184901,
      "balance_loss_mlp": 1.02066493,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.7321006890779396,
      "language_loss": 0.78059614,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.80252248,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 3.824240207672119
    },
    {
      "auxiliary_loss_clip": 0.01172503,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.05583835,
      "balance_loss_mlp": 1.02279258,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 3.7899315602192716,
      "language_loss": 0.57245427,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59449798,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 3.6897990703582764
    },
    {
      "auxiliary_loss_clip": 0.01181083,
      "auxiliary_loss_mlp": 0.01028274,
      "balance_loss_clip": 1.0548414,
      "balance_loss_mlp": 1.0199945,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 1.8005383874305272,
      "language_loss": 0.6981355,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.72022915,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.7074525356292725
    },
    {
      "auxiliary_loss_clip": 0.01178936,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 1.0517199,
      "balance_loss_mlp": 1.01739228,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.283417936717155,
      "language_loss": 0.76725328,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.78929484,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.6999731063842773
    },
    {
      "auxiliary_loss_clip": 0.01172498,
      "auxiliary_loss_mlp": 0.01034193,
      "balance_loss_clip": 1.05450428,
      "balance_loss_mlp": 1.02574098,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.8944383571366985,
      "language_loss": 0.91876906,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.94083595,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.66796612739563
    },
    {
      "auxiliary_loss_clip": 0.0118009,
      "auxiliary_loss_mlp": 0.01022678,
      "balance_loss_clip": 1.05777979,
      "balance_loss_mlp": 1.01457191,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.8693475091083067,
      "language_loss": 0.74843675,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77046448,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.68151593208313
    },
    {
      "auxiliary_loss_clip": 0.01162307,
      "auxiliary_loss_mlp": 0.01024623,
      "balance_loss_clip": 1.05287993,
      "balance_loss_mlp": 1.01630211,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.2467193232194513,
      "language_loss": 0.75598717,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.77785647,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 3.6718969345092773
    },
    {
      "auxiliary_loss_clip": 0.01165393,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.05608058,
      "balance_loss_mlp": 1.02499104,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 1.994253071178951,
      "language_loss": 0.85545397,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87744927,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.690800428390503
    },
    {
      "auxiliary_loss_clip": 0.01197227,
      "auxiliary_loss_mlp": 0.01030662,
      "balance_loss_clip": 1.05892277,
      "balance_loss_mlp": 1.02221656,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.3057629865616134,
      "language_loss": 0.73563159,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.75791049,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.6236178874969482
    },
    {
      "auxiliary_loss_clip": 0.01187521,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.05528152,
      "balance_loss_mlp": 1.01758528,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.197645489558984,
      "language_loss": 0.72010773,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74224651,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.5850682258605957
    },
    {
      "auxiliary_loss_clip": 0.01178868,
      "auxiliary_loss_mlp": 0.01033529,
      "balance_loss_clip": 1.05586314,
      "balance_loss_mlp": 1.02516699,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 1.9974773927744178,
      "language_loss": 0.73427975,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.7564038,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.844275951385498
    },
    {
      "auxiliary_loss_clip": 0.011509,
      "auxiliary_loss_mlp": 0.01033124,
      "balance_loss_clip": 1.05075073,
      "balance_loss_mlp": 1.02485085,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.864034963583854,
      "language_loss": 0.73182982,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.7536701,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.697741746902466
    },
    {
      "auxiliary_loss_clip": 0.01141671,
      "auxiliary_loss_mlp": 0.01029938,
      "balance_loss_clip": 1.04457736,
      "balance_loss_mlp": 1.02114654,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.6975156527918092,
      "language_loss": 0.81874681,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.84046286,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.7486989498138428
    },
    {
      "auxiliary_loss_clip": 0.01186601,
      "auxiliary_loss_mlp": 0.0102999,
      "balance_loss_clip": 1.05623507,
      "balance_loss_mlp": 1.02178824,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 2.1557671918434496,
      "language_loss": 0.73087263,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.75303853,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.659606456756592
    },
    {
      "auxiliary_loss_clip": 0.0117364,
      "auxiliary_loss_mlp": 0.01027553,
      "balance_loss_clip": 1.05391479,
      "balance_loss_mlp": 1.01955426,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 2.149023380333237,
      "language_loss": 0.84448147,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86649346,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.6668341159820557
    },
    {
      "auxiliary_loss_clip": 0.01196987,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.05675542,
      "balance_loss_mlp": 1.01783371,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.1950416487018822,
      "language_loss": 0.77050024,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.7927345,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 2.5456302165985107
    },
    {
      "auxiliary_loss_clip": 0.01184765,
      "auxiliary_loss_mlp": 0.00901718,
      "balance_loss_clip": 1.05948734,
      "balance_loss_mlp": 1.00184155,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.216717467694527,
      "language_loss": 0.82189977,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84276456,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 2.656001329421997
    },
    {
      "auxiliary_loss_clip": 0.01162619,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.05101895,
      "balance_loss_mlp": 1.01920938,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 1.7819856172596187,
      "language_loss": 0.73053873,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.7524389,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.7806732654571533
    },
    {
      "auxiliary_loss_clip": 0.01183775,
      "auxiliary_loss_mlp": 0.01028584,
      "balance_loss_clip": 1.05585039,
      "balance_loss_mlp": 1.02007854,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 3.1017678903086394,
      "language_loss": 0.81315696,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83528054,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.66687273979187
    },
    {
      "auxiliary_loss_clip": 0.01153839,
      "auxiliary_loss_mlp": 0.00901941,
      "balance_loss_clip": 1.05215454,
      "balance_loss_mlp": 1.00180447,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 1.7904528771461645,
      "language_loss": 0.87362713,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89418495,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.7960546016693115
    },
    {
      "auxiliary_loss_clip": 0.01132919,
      "auxiliary_loss_mlp": 0.01026788,
      "balance_loss_clip": 1.04573393,
      "balance_loss_mlp": 1.01882529,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.983444233045505,
      "language_loss": 0.88707566,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90867281,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.784432888031006
    },
    {
      "auxiliary_loss_clip": 0.01172261,
      "auxiliary_loss_mlp": 0.01030411,
      "balance_loss_clip": 1.05117965,
      "balance_loss_mlp": 1.02140474,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 3.563958691138208,
      "language_loss": 0.80753243,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82955921,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.72023868560791
    },
    {
      "auxiliary_loss_clip": 0.01182548,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.05833769,
      "balance_loss_mlp": 1.02586007,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.7155718734874026,
      "language_loss": 0.85997719,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88214302,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.611443519592285
    },
    {
      "auxiliary_loss_clip": 0.01170966,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.04914331,
      "balance_loss_mlp": 1.02083635,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.9545555517301012,
      "language_loss": 0.90211868,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92413092,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 3.690201997756958
    },
    {
      "auxiliary_loss_clip": 0.0116831,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.05061507,
      "balance_loss_mlp": 1.02044106,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.6627650098867264,
      "language_loss": 0.69832832,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.72030187,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.619924783706665
    },
    {
      "auxiliary_loss_clip": 0.01155029,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.04645395,
      "balance_loss_mlp": 1.01952839,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 3.033430986287708,
      "language_loss": 0.79935157,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.82118326,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.958339214324951
    },
    {
      "auxiliary_loss_clip": 0.01181764,
      "auxiliary_loss_mlp": 0.01026106,
      "balance_loss_clip": 1.05802286,
      "balance_loss_mlp": 1.01803529,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.3915017869649433,
      "language_loss": 0.67677087,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69884956,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.760246753692627
    },
    {
      "auxiliary_loss_clip": 0.01178029,
      "auxiliary_loss_mlp": 0.01024152,
      "balance_loss_clip": 1.05185485,
      "balance_loss_mlp": 1.01621866,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.1264441342295535,
      "language_loss": 0.79015201,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.81217384,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 4.525794744491577
    },
    {
      "auxiliary_loss_clip": 0.01162514,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.05674529,
      "balance_loss_mlp": 1.01860523,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.7823593872299803,
      "language_loss": 0.79533899,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.81723034,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.691838502883911
    },
    {
      "auxiliary_loss_clip": 0.01200525,
      "auxiliary_loss_mlp": 0.0103195,
      "balance_loss_clip": 1.06037188,
      "balance_loss_mlp": 1.02284789,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.3876121539669786,
      "language_loss": 0.74891472,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77123946,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.603238821029663
    },
    {
      "auxiliary_loss_clip": 0.01158515,
      "auxiliary_loss_mlp": 0.01027183,
      "balance_loss_clip": 1.0543716,
      "balance_loss_mlp": 1.01913655,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.6980201150317813,
      "language_loss": 0.76565409,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78751111,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.801060438156128
    },
    {
      "auxiliary_loss_clip": 0.01170588,
      "auxiliary_loss_mlp": 0.01033921,
      "balance_loss_clip": 1.05480933,
      "balance_loss_mlp": 1.02591002,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.7041728529601865,
      "language_loss": 0.71591473,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73795986,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.791436195373535
    },
    {
      "auxiliary_loss_clip": 0.01198688,
      "auxiliary_loss_mlp": 0.0103496,
      "balance_loss_clip": 1.06082153,
      "balance_loss_mlp": 1.02623427,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 2.205968332823902,
      "language_loss": 0.81231105,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83464748,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 3.5727033615112305
    },
    {
      "auxiliary_loss_clip": 0.01185252,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05807066,
      "balance_loss_mlp": 1.02166367,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.6943531507125265,
      "language_loss": 0.75753343,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.77968681,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.7122697830200195
    },
    {
      "auxiliary_loss_clip": 0.01141905,
      "auxiliary_loss_mlp": 0.00902194,
      "balance_loss_clip": 1.05441117,
      "balance_loss_mlp": 1.00161219,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 2.120698197950486,
      "language_loss": 0.80143869,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82187968,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.702176809310913
    },
    {
      "auxiliary_loss_clip": 0.0116009,
      "auxiliary_loss_mlp": 0.01028988,
      "balance_loss_clip": 1.04916596,
      "balance_loss_mlp": 1.02103102,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.533070625418457,
      "language_loss": 0.83721161,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85910237,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.73394775390625
    },
    {
      "auxiliary_loss_clip": 0.01178364,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.05493069,
      "balance_loss_mlp": 1.01749003,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.7089759445376669,
      "language_loss": 0.8231886,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84523213,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.6724648475646973
    },
    {
      "auxiliary_loss_clip": 0.01177292,
      "auxiliary_loss_mlp": 0.01030215,
      "balance_loss_clip": 1.05823922,
      "balance_loss_mlp": 1.02178049,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 1.9675752966593878,
      "language_loss": 0.70404506,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72612011,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.61663818359375
    },
    {
      "auxiliary_loss_clip": 0.01173569,
      "auxiliary_loss_mlp": 0.01030557,
      "balance_loss_clip": 1.05566669,
      "balance_loss_mlp": 1.02252221,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 2.064369601832406,
      "language_loss": 0.82502431,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.84706557,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.7428324222564697
    },
    {
      "auxiliary_loss_clip": 0.01184568,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.05664575,
      "balance_loss_mlp": 1.02164376,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.3663560365920007,
      "language_loss": 0.75525099,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.77739859,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.6477129459381104
    },
    {
      "auxiliary_loss_clip": 0.0119885,
      "auxiliary_loss_mlp": 0.01029854,
      "balance_loss_clip": 1.06002426,
      "balance_loss_mlp": 1.02092528,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.9037462210056348,
      "language_loss": 0.84588617,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86817324,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.634190797805786
    },
    {
      "auxiliary_loss_clip": 0.01186657,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.05768609,
      "balance_loss_mlp": 1.01834941,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 2.622803231061759,
      "language_loss": 0.71712166,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73925549,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.675227642059326
    },
    {
      "auxiliary_loss_clip": 0.01187303,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.05928409,
      "balance_loss_mlp": 1.02461052,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 2.601773645053024,
      "language_loss": 0.76539135,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78759342,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.6773786544799805
    },
    {
      "auxiliary_loss_clip": 0.01158575,
      "auxiliary_loss_mlp": 0.00902768,
      "balance_loss_clip": 1.04897809,
      "balance_loss_mlp": 1.00186467,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.6674255963069013,
      "language_loss": 0.78236371,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80297714,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.6862220764160156
    },
    {
      "auxiliary_loss_clip": 0.01181026,
      "auxiliary_loss_mlp": 0.01032837,
      "balance_loss_clip": 1.05620313,
      "balance_loss_mlp": 1.02476704,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 2.794660575042361,
      "language_loss": 0.70125484,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72339344,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.75704288482666
    },
    {
      "auxiliary_loss_clip": 0.01158589,
      "auxiliary_loss_mlp": 0.01025653,
      "balance_loss_clip": 1.05204928,
      "balance_loss_mlp": 1.01711154,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.71555252004897,
      "language_loss": 0.63816643,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66000885,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.696424961090088
    },
    {
      "auxiliary_loss_clip": 0.01160785,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.05114508,
      "balance_loss_mlp": 1.02214408,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 2.7973927666914324,
      "language_loss": 0.76671767,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78863311,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.720721960067749
    },
    {
      "auxiliary_loss_clip": 0.01175791,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.05519545,
      "balance_loss_mlp": 1.02229702,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.530363065593084,
      "language_loss": 0.66626418,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68833071,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.670024871826172
    },
    {
      "auxiliary_loss_clip": 0.01198245,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.05944395,
      "balance_loss_mlp": 1.02229726,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 2.077344779067672,
      "language_loss": 0.81770426,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.8399936,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.6286089420318604
    },
    {
      "auxiliary_loss_clip": 0.01187865,
      "auxiliary_loss_mlp": 0.01033889,
      "balance_loss_clip": 1.05891657,
      "balance_loss_mlp": 1.02446568,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.0054951388032842,
      "language_loss": 0.80545628,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82767385,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 3.622464656829834
    },
    {
      "auxiliary_loss_clip": 0.01187343,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.0578959,
      "balance_loss_mlp": 1.02598524,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.7452797515736969,
      "language_loss": 0.65219748,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67442298,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.6696388721466064
    },
    {
      "auxiliary_loss_clip": 0.01197051,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.05830884,
      "balance_loss_mlp": 1.02145123,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 1.9871407769398892,
      "language_loss": 0.85269606,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87497187,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.5746850967407227
    },
    {
      "auxiliary_loss_clip": 0.0115418,
      "auxiliary_loss_mlp": 0.00902112,
      "balance_loss_clip": 1.05316806,
      "balance_loss_mlp": 1.0016315,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.1042670244178963,
      "language_loss": 0.78205907,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80262196,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.681044340133667
    },
    {
      "auxiliary_loss_clip": 0.01167434,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.05061567,
      "balance_loss_mlp": 1.02134252,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.94557241366284,
      "language_loss": 0.78457087,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80653846,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 3.6449244022369385
    },
    {
      "auxiliary_loss_clip": 0.01198934,
      "auxiliary_loss_mlp": 0.01031042,
      "balance_loss_clip": 1.05952406,
      "balance_loss_mlp": 1.02247047,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 3.1771612374817866,
      "language_loss": 0.74427879,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.7665785,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 3.5291435718536377
    },
    {
      "auxiliary_loss_clip": 0.01158646,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05699301,
      "balance_loss_mlp": 1.02073705,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 23.74899188772889,
      "language_loss": 0.68206561,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70395064,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.7228221893310547
    },
    {
      "auxiliary_loss_clip": 0.01183,
      "auxiliary_loss_mlp": 0.01031579,
      "balance_loss_clip": 1.05888176,
      "balance_loss_mlp": 1.02325261,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.5511145807058455,
      "language_loss": 0.82623148,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84837723,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.6942594051361084
    },
    {
      "auxiliary_loss_clip": 0.01198546,
      "auxiliary_loss_mlp": 0.01034135,
      "balance_loss_clip": 1.05857539,
      "balance_loss_mlp": 1.02536106,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.244499906083836,
      "language_loss": 0.74350131,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76582813,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.5646514892578125
    },
    {
      "auxiliary_loss_clip": 0.0118903,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.05679274,
      "balance_loss_mlp": 1.01984656,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 2.1506540148627824,
      "language_loss": 0.68498129,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.7071588,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.68926739692688
    },
    {
      "auxiliary_loss_clip": 0.011791,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.05822229,
      "balance_loss_mlp": 1.0175972,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.9120897116809423,
      "language_loss": 0.8428185,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86487067,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 3.6565675735473633
    },
    {
      "auxiliary_loss_clip": 0.01188254,
      "auxiliary_loss_mlp": 0.01033862,
      "balance_loss_clip": 1.05860376,
      "balance_loss_mlp": 1.02464128,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.8078244021576118,
      "language_loss": 0.81996167,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84218276,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.6120669841766357
    },
    {
      "auxiliary_loss_clip": 0.01144008,
      "auxiliary_loss_mlp": 0.00902613,
      "balance_loss_clip": 1.05021286,
      "balance_loss_mlp": 1.00169075,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 3.110920823660856,
      "language_loss": 0.82505882,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.84552503,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.8638672828674316
    },
    {
      "auxiliary_loss_clip": 0.01165855,
      "auxiliary_loss_mlp": 0.01033522,
      "balance_loss_clip": 1.05274034,
      "balance_loss_mlp": 1.02441454,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 1.9027588207288997,
      "language_loss": 0.80258894,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.8245827,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.6002511978149414
    },
    {
      "auxiliary_loss_clip": 0.01192439,
      "auxiliary_loss_mlp": 0.00902654,
      "balance_loss_clip": 1.05645597,
      "balance_loss_mlp": 1.00169802,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.8574534601677484,
      "language_loss": 0.76967549,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.79062647,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.7290735244750977
    },
    {
      "auxiliary_loss_clip": 0.01101303,
      "auxiliary_loss_mlp": 0.0101351,
      "balance_loss_clip": 1.02792692,
      "balance_loss_mlp": 1.01172221,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.9472830762911378,
      "language_loss": 0.60031688,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62146497,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.0724711418151855
    },
    {
      "auxiliary_loss_clip": 0.01177388,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.06063712,
      "balance_loss_mlp": 1.02201104,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 2.141261496694445,
      "language_loss": 0.63923693,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66131318,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.653928518295288
    },
    {
      "auxiliary_loss_clip": 0.01197924,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.05852842,
      "balance_loss_mlp": 1.02010274,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 2.462769295157232,
      "language_loss": 0.70023739,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.7225045,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.6710758209228516
    },
    {
      "auxiliary_loss_clip": 0.01178734,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.0578469,
      "balance_loss_mlp": 1.0185225,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 2.644961813408088,
      "language_loss": 0.70076114,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72282296,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.6567869186401367
    },
    {
      "auxiliary_loss_clip": 0.01165631,
      "auxiliary_loss_mlp": 0.0103813,
      "balance_loss_clip": 1.05483353,
      "balance_loss_mlp": 1.02881932,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.192663505516212,
      "language_loss": 0.86985868,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.89189625,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 2.6897151470184326
    },
    {
      "auxiliary_loss_clip": 0.0118462,
      "auxiliary_loss_mlp": 0.01029179,
      "balance_loss_clip": 1.05782497,
      "balance_loss_mlp": 1.02031529,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 2.1512027962000273,
      "language_loss": 0.77604008,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79817808,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.595864772796631
    },
    {
      "auxiliary_loss_clip": 0.01192049,
      "auxiliary_loss_mlp": 0.01041298,
      "balance_loss_clip": 1.0573225,
      "balance_loss_mlp": 1.03252983,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 2.0302376328960134,
      "language_loss": 0.722471,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74480456,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.6502015590667725
    },
    {
      "auxiliary_loss_clip": 0.01088426,
      "auxiliary_loss_mlp": 0.00891911,
      "balance_loss_clip": 1.02268982,
      "balance_loss_mlp": 0.99987745,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0187496565918785,
      "language_loss": 0.6407057,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66050911,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.2537832260131836
    },
    {
      "auxiliary_loss_clip": 0.01136936,
      "auxiliary_loss_mlp": 0.01030002,
      "balance_loss_clip": 1.04685664,
      "balance_loss_mlp": 1.02024496,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.4983933917167156,
      "language_loss": 0.78411043,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80577981,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.8200089931488037
    },
    {
      "auxiliary_loss_clip": 0.01149671,
      "auxiliary_loss_mlp": 0.01037145,
      "balance_loss_clip": 1.05089688,
      "balance_loss_mlp": 1.02822232,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 2.2363288415161207,
      "language_loss": 0.82614666,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.84801483,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.768366575241089
    },
    {
      "auxiliary_loss_clip": 0.01180106,
      "auxiliary_loss_mlp": 0.01034372,
      "balance_loss_clip": 1.05635023,
      "balance_loss_mlp": 1.02540803,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.8578993244643003,
      "language_loss": 0.7625488,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.7846936,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.7559428215026855
    },
    {
      "auxiliary_loss_clip": 0.01192412,
      "auxiliary_loss_mlp": 0.01034165,
      "balance_loss_clip": 1.05626965,
      "balance_loss_mlp": 1.02505779,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.7977646756871604,
      "language_loss": 0.7650947,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78736049,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.60211443901062
    },
    {
      "auxiliary_loss_clip": 0.01150371,
      "auxiliary_loss_mlp": 0.01032223,
      "balance_loss_clip": 1.05612803,
      "balance_loss_mlp": 1.02313876,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 1.9555561658383132,
      "language_loss": 0.92005503,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94188094,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 3.645756244659424
    },
    {
      "auxiliary_loss_clip": 0.01154108,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.05082464,
      "balance_loss_mlp": 1.01968348,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 2.353499866191278,
      "language_loss": 0.77572167,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79754609,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.7213680744171143
    },
    {
      "auxiliary_loss_clip": 0.01178472,
      "auxiliary_loss_mlp": 0.01034271,
      "balance_loss_clip": 1.05410624,
      "balance_loss_mlp": 1.02525878,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.9099563893192437,
      "language_loss": 0.76710999,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78923738,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.6352932453155518
    },
    {
      "auxiliary_loss_clip": 0.01172425,
      "auxiliary_loss_mlp": 0.01026431,
      "balance_loss_clip": 1.05576277,
      "balance_loss_mlp": 1.01781273,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.8252286895072471,
      "language_loss": 0.70205235,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72404087,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.735567092895508
    },
    {
      "auxiliary_loss_clip": 0.01152283,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.05060935,
      "balance_loss_mlp": 1.01811242,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.9095994351677847,
      "language_loss": 0.78747094,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.80926448,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 4.60422420501709
    },
    {
      "auxiliary_loss_clip": 0.01184788,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.05342603,
      "balance_loss_mlp": 1.02189589,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 2.0971330299231923,
      "language_loss": 0.79917186,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.82132518,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.690751314163208
    },
    {
      "auxiliary_loss_clip": 0.01075078,
      "auxiliary_loss_mlp": 0.01007074,
      "balance_loss_clip": 1.02047753,
      "balance_loss_mlp": 1.00532198,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7219398067108133,
      "language_loss": 0.56463283,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58545434,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.2261741161346436
    },
    {
      "auxiliary_loss_clip": 0.01158672,
      "auxiliary_loss_mlp": 0.01030625,
      "balance_loss_clip": 1.05271924,
      "balance_loss_mlp": 1.02195859,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 1.9730947846912106,
      "language_loss": 0.77529687,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.79718983,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.7164134979248047
    },
    {
      "auxiliary_loss_clip": 0.01183058,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.05501747,
      "balance_loss_mlp": 1.02135634,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 2.065454672458385,
      "language_loss": 0.75558352,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77771425,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.717500686645508
    },
    {
      "auxiliary_loss_clip": 0.01078953,
      "auxiliary_loss_mlp": 0.0100289,
      "balance_loss_clip": 1.01874542,
      "balance_loss_mlp": 1.00106657,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8428215676562198,
      "language_loss": 0.60949647,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.63031495,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 4.184901475906372
    },
    {
      "auxiliary_loss_clip": 0.01196902,
      "auxiliary_loss_mlp": 0.01028192,
      "balance_loss_clip": 1.0588311,
      "balance_loss_mlp": 1.01918018,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 2.0252742925993057,
      "language_loss": 0.66692078,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.68917173,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.592137336730957
    },
    {
      "auxiliary_loss_clip": 0.01173895,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.05101871,
      "balance_loss_mlp": 1.0248152,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.7726156293066868,
      "language_loss": 0.79874891,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.8208276,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.722334861755371
    },
    {
      "auxiliary_loss_clip": 0.01193056,
      "auxiliary_loss_mlp": 0.0102856,
      "balance_loss_clip": 1.05926704,
      "balance_loss_mlp": 1.01948202,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.1610285002469904,
      "language_loss": 0.86242008,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88463628,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.561122179031372
    },
    {
      "auxiliary_loss_clip": 0.01175106,
      "auxiliary_loss_mlp": 0.0102864,
      "balance_loss_clip": 1.0536412,
      "balance_loss_mlp": 1.01913869,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 2.217144928059568,
      "language_loss": 0.81829596,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.8403334,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.662572145462036
    },
    {
      "auxiliary_loss_clip": 0.01162817,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.04993367,
      "balance_loss_mlp": 1.02292967,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 4.155615830817118,
      "language_loss": 0.75864106,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.78058797,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.664149761199951
    },
    {
      "auxiliary_loss_clip": 0.01090449,
      "auxiliary_loss_mlp": 0.01002034,
      "balance_loss_clip": 1.01787663,
      "balance_loss_mlp": 1.00022221,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9919120295760212,
      "language_loss": 0.62381899,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64474386,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.057319164276123
    },
    {
      "auxiliary_loss_clip": 0.01200215,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.06440949,
      "balance_loss_mlp": 1.02116585,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.7091772258154356,
      "language_loss": 0.8142153,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.83650887,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.5737099647521973
    },
    {
      "auxiliary_loss_clip": 0.01178019,
      "auxiliary_loss_mlp": 0.01028479,
      "balance_loss_clip": 1.05691743,
      "balance_loss_mlp": 1.01961637,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 1.9549631598843453,
      "language_loss": 0.74170399,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76376903,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.6649885177612305
    },
    {
      "auxiliary_loss_clip": 0.01156652,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.05083847,
      "balance_loss_mlp": 1.01593363,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 3.003889575963834,
      "language_loss": 0.79397041,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81577992,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.7162225246429443
    },
    {
      "auxiliary_loss_clip": 0.01157966,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.05480409,
      "balance_loss_mlp": 1.01783371,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.7100524832592947,
      "language_loss": 0.88892853,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.9107765,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.7067112922668457
    },
    {
      "auxiliary_loss_clip": 0.01172607,
      "auxiliary_loss_mlp": 0.01031257,
      "balance_loss_clip": 1.05401015,
      "balance_loss_mlp": 1.02306795,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.8362258010815733,
      "language_loss": 0.68705678,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70909536,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.753584384918213
    },
    {
      "auxiliary_loss_clip": 0.01177537,
      "auxiliary_loss_mlp": 0.01031946,
      "balance_loss_clip": 1.05962658,
      "balance_loss_mlp": 1.02304721,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 3.2305811842712733,
      "language_loss": 0.75210333,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77419817,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.6195285320281982
    },
    {
      "auxiliary_loss_clip": 0.01144526,
      "auxiliary_loss_mlp": 0.01029496,
      "balance_loss_clip": 1.04750133,
      "balance_loss_mlp": 1.02062678,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.5531372001873036,
      "language_loss": 0.74826086,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77000105,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.7272558212280273
    },
    {
      "auxiliary_loss_clip": 0.01174476,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.0566864,
      "balance_loss_mlp": 1.02427852,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 2.016292556457696,
      "language_loss": 0.77564776,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79772252,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.624018907546997
    },
    {
      "auxiliary_loss_clip": 0.01191504,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.05831981,
      "balance_loss_mlp": 1.02144611,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.8656418350533066,
      "language_loss": 0.79592133,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81813639,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.673973321914673
    },
    {
      "auxiliary_loss_clip": 0.01170264,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.05389071,
      "balance_loss_mlp": 1.02155721,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.2079583094971484,
      "language_loss": 0.71520388,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.73720288,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.6739585399627686
    },
    {
      "auxiliary_loss_clip": 0.01171034,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.05351388,
      "balance_loss_mlp": 1.01824844,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.039309600020619,
      "language_loss": 0.70047367,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72245556,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 3.6674907207489014
    },
    {
      "auxiliary_loss_clip": 0.01175311,
      "auxiliary_loss_mlp": 0.01037564,
      "balance_loss_clip": 1.05176449,
      "balance_loss_mlp": 1.02809858,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.618040722615314,
      "language_loss": 0.86417341,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88630211,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.6323509216308594
    },
    {
      "auxiliary_loss_clip": 0.01051002,
      "auxiliary_loss_mlp": 0.01002686,
      "balance_loss_clip": 1.01540089,
      "balance_loss_mlp": 1.00087416,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8507998094423755,
      "language_loss": 0.5588603,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.5793972,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.369072198867798
    },
    {
      "auxiliary_loss_clip": 0.0115001,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 1.05024457,
      "balance_loss_mlp": 1.0198735,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.8815197437042779,
      "language_loss": 0.85046726,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87225026,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 3.0805091857910156
    },
    {
      "auxiliary_loss_clip": 0.01144264,
      "auxiliary_loss_mlp": 0.01030855,
      "balance_loss_clip": 1.04746795,
      "balance_loss_mlp": 1.02299285,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.6396314736101125,
      "language_loss": 0.76497537,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78672659,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 3.8282670974731445
    },
    {
      "auxiliary_loss_clip": 0.01169698,
      "auxiliary_loss_mlp": 0.01028708,
      "balance_loss_clip": 1.05511713,
      "balance_loss_mlp": 1.02023268,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.568906339322397,
      "language_loss": 0.73923576,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76121986,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 3.7850565910339355
    },
    {
      "auxiliary_loss_clip": 0.01173976,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.05598092,
      "balance_loss_mlp": 1.01824522,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 2.2383428845844473,
      "language_loss": 0.82667863,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84868503,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.7984542846679688
    },
    {
      "auxiliary_loss_clip": 0.01175235,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.05523038,
      "balance_loss_mlp": 1.01893091,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 1.9346666395271133,
      "language_loss": 0.75436306,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.77639288,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.7006335258483887
    },
    {
      "auxiliary_loss_clip": 0.01144349,
      "auxiliary_loss_mlp": 0.01029059,
      "balance_loss_clip": 1.04479718,
      "balance_loss_mlp": 1.02042854,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.564093657813499,
      "language_loss": 0.76620781,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78794181,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.771914005279541
    },
    {
      "auxiliary_loss_clip": 0.01175402,
      "auxiliary_loss_mlp": 0.01031806,
      "balance_loss_clip": 1.05124235,
      "balance_loss_mlp": 1.02245378,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.8125883008112122,
      "language_loss": 0.820351,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84242308,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 3.645315170288086
    },
    {
      "auxiliary_loss_clip": 0.01180685,
      "auxiliary_loss_mlp": 0.01033526,
      "balance_loss_clip": 1.05422997,
      "balance_loss_mlp": 1.02412677,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 2.055534937095234,
      "language_loss": 0.78043145,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80257356,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.715735912322998
    },
    {
      "auxiliary_loss_clip": 0.01183376,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.05729175,
      "balance_loss_mlp": 1.02598083,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.4036318683616624,
      "language_loss": 0.74878305,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.77095509,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.676006317138672
    },
    {
      "auxiliary_loss_clip": 0.01179187,
      "auxiliary_loss_mlp": 0.01032927,
      "balance_loss_clip": 1.05473733,
      "balance_loss_mlp": 1.02464199,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.6128593111074125,
      "language_loss": 0.67027301,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69239414,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.601027011871338
    },
    {
      "auxiliary_loss_clip": 0.01155092,
      "auxiliary_loss_mlp": 0.00902287,
      "balance_loss_clip": 1.04944479,
      "balance_loss_mlp": 1.00155902,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.5242893977027645,
      "language_loss": 0.69482893,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71540272,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.769228935241699
    },
    {
      "auxiliary_loss_clip": 0.01181341,
      "auxiliary_loss_mlp": 0.01023157,
      "balance_loss_clip": 1.0538857,
      "balance_loss_mlp": 1.01453233,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 2.171814037904447,
      "language_loss": 0.63119006,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65323508,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.7651288509368896
    },
    {
      "auxiliary_loss_clip": 0.01185071,
      "auxiliary_loss_mlp": 0.01030175,
      "balance_loss_clip": 1.05659485,
      "balance_loss_mlp": 1.02174056,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 1.987566015040937,
      "language_loss": 0.6849314,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.70708382,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.7922933101654053
    },
    {
      "auxiliary_loss_clip": 0.01174053,
      "auxiliary_loss_mlp": 0.01031534,
      "balance_loss_clip": 1.053352,
      "balance_loss_mlp": 1.02253926,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.3536752890828927,
      "language_loss": 0.76256859,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78462446,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.6282975673675537
    },
    {
      "auxiliary_loss_clip": 0.01184756,
      "auxiliary_loss_mlp": 0.01024246,
      "balance_loss_clip": 1.05530763,
      "balance_loss_mlp": 1.01640224,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 2.0894779537917816,
      "language_loss": 0.87869644,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.90078652,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.667600154876709
    },
    {
      "auxiliary_loss_clip": 0.01196253,
      "auxiliary_loss_mlp": 0.01030887,
      "balance_loss_clip": 1.05742848,
      "balance_loss_mlp": 1.02247667,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.1578139720616556,
      "language_loss": 0.8027696,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82504106,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.602259397506714
    },
    {
      "auxiliary_loss_clip": 0.01185191,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.0539639,
      "balance_loss_mlp": 1.01975775,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 3.1427769385316653,
      "language_loss": 0.85736972,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.87950414,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.616007089614868
    },
    {
      "auxiliary_loss_clip": 0.01191785,
      "auxiliary_loss_mlp": 0.0102876,
      "balance_loss_clip": 1.05537558,
      "balance_loss_mlp": 1.01998019,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.878678424927716,
      "language_loss": 0.82007957,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.84228498,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.5933895111083984
    },
    {
      "auxiliary_loss_clip": 0.01184647,
      "auxiliary_loss_mlp": 0.01034046,
      "balance_loss_clip": 1.05623579,
      "balance_loss_mlp": 1.02542686,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.7362333071324094,
      "language_loss": 0.75693202,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.7791189,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.6459708213806152
    },
    {
      "auxiliary_loss_clip": 0.01087274,
      "auxiliary_loss_mlp": 0.01003769,
      "balance_loss_clip": 1.02059579,
      "balance_loss_mlp": 1.00219584,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.982814603937376,
      "language_loss": 0.59195876,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.6128692,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.1052284240722656
    },
    {
      "auxiliary_loss_clip": 0.01169402,
      "auxiliary_loss_mlp": 0.01033874,
      "balance_loss_clip": 1.05237269,
      "balance_loss_mlp": 1.02431321,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 2.1067805782964832,
      "language_loss": 0.76626819,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78830093,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.7125048637390137
    },
    {
      "auxiliary_loss_clip": 0.01167725,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.05243611,
      "balance_loss_mlp": 1.02188575,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.8841564538714342,
      "language_loss": 0.66056335,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68254733,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.7730445861816406
    },
    {
      "auxiliary_loss_clip": 0.01190282,
      "auxiliary_loss_mlp": 0.01036768,
      "balance_loss_clip": 1.05684578,
      "balance_loss_mlp": 1.0286026,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.8479924635497904,
      "language_loss": 0.81269479,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83496535,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.67948579788208
    },
    {
      "auxiliary_loss_clip": 0.01176253,
      "auxiliary_loss_mlp": 0.01034324,
      "balance_loss_clip": 1.05524874,
      "balance_loss_mlp": 1.0257647,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.3982397891986937,
      "language_loss": 0.7550627,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77716851,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 3.570023775100708
    },
    {
      "auxiliary_loss_clip": 0.01198253,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.05802703,
      "balance_loss_mlp": 1.02379775,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 2.6610905581117748,
      "language_loss": 0.84425682,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.8665666,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.5994725227355957
    },
    {
      "auxiliary_loss_clip": 0.01169647,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.05085838,
      "balance_loss_mlp": 1.02004921,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.657002785432834,
      "language_loss": 0.78988373,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.81186295,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6390299797058105
    },
    {
      "auxiliary_loss_clip": 0.01165266,
      "auxiliary_loss_mlp": 0.01024427,
      "balance_loss_clip": 1.05532634,
      "balance_loss_mlp": 1.01589119,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.777655288205453,
      "language_loss": 0.69858158,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.72047848,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.7654166221618652
    },
    {
      "auxiliary_loss_clip": 0.01180526,
      "auxiliary_loss_mlp": 0.01034545,
      "balance_loss_clip": 1.0565052,
      "balance_loss_mlp": 1.02593184,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.0588202641709024,
      "language_loss": 0.87079263,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89294338,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.635521650314331
    },
    {
      "auxiliary_loss_clip": 0.01192219,
      "auxiliary_loss_mlp": 0.01031644,
      "balance_loss_clip": 1.06195211,
      "balance_loss_mlp": 1.02174377,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.201551539363506,
      "language_loss": 0.78712094,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80935955,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 3.5406365394592285
    },
    {
      "auxiliary_loss_clip": 0.01173315,
      "auxiliary_loss_mlp": 0.01035824,
      "balance_loss_clip": 1.05623865,
      "balance_loss_mlp": 1.02712154,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 3.4615899716832867,
      "language_loss": 0.83178997,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.85388136,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.668792247772217
    },
    {
      "auxiliary_loss_clip": 0.0119652,
      "auxiliary_loss_mlp": 0.01027651,
      "balance_loss_clip": 1.05978274,
      "balance_loss_mlp": 1.01940143,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 4.139817579249285,
      "language_loss": 0.79544544,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81768715,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.6078548431396484
    },
    {
      "auxiliary_loss_clip": 0.01173881,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.05957305,
      "balance_loss_mlp": 1.01859522,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 2.277002734444474,
      "language_loss": 0.74454558,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76655686,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.707686424255371
    },
    {
      "auxiliary_loss_clip": 0.01196835,
      "auxiliary_loss_mlp": 0.01027876,
      "balance_loss_clip": 1.05790126,
      "balance_loss_mlp": 1.01903677,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.9471383420924273,
      "language_loss": 0.8120321,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83427918,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.5950183868408203
    },
    {
      "auxiliary_loss_clip": 0.01194343,
      "auxiliary_loss_mlp": 0.01034464,
      "balance_loss_clip": 1.05776286,
      "balance_loss_mlp": 1.02539802,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 2.063024660651035,
      "language_loss": 0.80738938,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82967746,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 3.5913336277008057
    },
    {
      "auxiliary_loss_clip": 0.01189765,
      "auxiliary_loss_mlp": 0.00902742,
      "balance_loss_clip": 1.05797839,
      "balance_loss_mlp": 1.00121343,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.464959285670256,
      "language_loss": 0.85802668,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.87895173,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.6453490257263184
    },
    {
      "auxiliary_loss_clip": 0.01167186,
      "auxiliary_loss_mlp": 0.01028685,
      "balance_loss_clip": 1.05734849,
      "balance_loss_mlp": 1.01988184,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 3.3774475806250814,
      "language_loss": 0.79640841,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81836712,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.831413745880127
    },
    {
      "auxiliary_loss_clip": 0.01177999,
      "auxiliary_loss_mlp": 0.00902385,
      "balance_loss_clip": 1.0554291,
      "balance_loss_mlp": 1.00113738,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.6304558845258694,
      "language_loss": 0.75241572,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77321953,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.78299617767334
    },
    {
      "auxiliary_loss_clip": 0.01086581,
      "auxiliary_loss_mlp": 0.01003537,
      "balance_loss_clip": 1.01933289,
      "balance_loss_mlp": 1.00202882,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8782923909132795,
      "language_loss": 0.65222669,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67312789,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.187094211578369
    },
    {
      "auxiliary_loss_clip": 0.01164593,
      "auxiliary_loss_mlp": 0.00902338,
      "balance_loss_clip": 1.05170047,
      "balance_loss_mlp": 1.00107837,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.4824293906952482,
      "language_loss": 0.85554934,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87621856,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.744251251220703
    },
    {
      "auxiliary_loss_clip": 0.01169969,
      "auxiliary_loss_mlp": 0.01034419,
      "balance_loss_clip": 1.05514097,
      "balance_loss_mlp": 1.02525759,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.8171241547952308,
      "language_loss": 0.86427134,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88631523,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.660874605178833
    },
    {
      "auxiliary_loss_clip": 0.01188375,
      "auxiliary_loss_mlp": 0.00903021,
      "balance_loss_clip": 1.05739307,
      "balance_loss_mlp": 1.00116861,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.9765243231792557,
      "language_loss": 0.77463293,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79554689,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.6369428634643555
    },
    {
      "auxiliary_loss_clip": 0.01183639,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.05814302,
      "balance_loss_mlp": 1.0177182,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.433729270981714,
      "language_loss": 0.70210302,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.72419751,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.7515501976013184
    },
    {
      "auxiliary_loss_clip": 0.01199843,
      "auxiliary_loss_mlp": 0.00902706,
      "balance_loss_clip": 1.05972409,
      "balance_loss_mlp": 1.00099051,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 1.683547922114172,
      "language_loss": 0.72452164,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74554706,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.6051361560821533
    },
    {
      "auxiliary_loss_clip": 0.01170833,
      "auxiliary_loss_mlp": 0.00902469,
      "balance_loss_clip": 1.05189037,
      "balance_loss_mlp": 1.00118256,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.1069554028870088,
      "language_loss": 0.83317709,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85391009,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.6707370281219482
    },
    {
      "auxiliary_loss_clip": 0.01180372,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.05349422,
      "balance_loss_mlp": 1.02114558,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.9174604363070764,
      "language_loss": 0.78893638,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81103295,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.6528055667877197
    },
    {
      "auxiliary_loss_clip": 0.0119466,
      "auxiliary_loss_mlp": 0.01028856,
      "balance_loss_clip": 1.05561686,
      "balance_loss_mlp": 1.02044606,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 1.946010502216186,
      "language_loss": 0.71529168,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.73752689,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.618295192718506
    },
    {
      "auxiliary_loss_clip": 0.01190755,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.05579102,
      "balance_loss_mlp": 1.02118111,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.4956915437765956,
      "language_loss": 0.84267867,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86488414,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.6016509532928467
    },
    {
      "auxiliary_loss_clip": 0.01195673,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.05685449,
      "balance_loss_mlp": 1.01907539,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.3926868217315405,
      "language_loss": 0.780936,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.8031714,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.602552652359009
    },
    {
      "auxiliary_loss_clip": 0.01183255,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.05520058,
      "balance_loss_mlp": 1.01660824,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 1.7790892858345833,
      "language_loss": 0.83628064,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85835421,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.654144525527954
    },
    {
      "auxiliary_loss_clip": 0.01172455,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.05444455,
      "balance_loss_mlp": 1.01831722,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 2.235912831218347,
      "language_loss": 0.73158085,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75357693,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.7693846225738525
    },
    {
      "auxiliary_loss_clip": 0.01177241,
      "auxiliary_loss_mlp": 0.0090265,
      "balance_loss_clip": 1.05584598,
      "balance_loss_mlp": 1.00111997,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.573018365757647,
      "language_loss": 0.89699829,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.91779721,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 3.5916075706481934
    },
    {
      "auxiliary_loss_clip": 0.01089985,
      "auxiliary_loss_mlp": 0.01005263,
      "balance_loss_clip": 1.0198704,
      "balance_loss_mlp": 1.00374305,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8758051803095901,
      "language_loss": 0.66115731,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68210977,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.155432939529419
    },
    {
      "auxiliary_loss_clip": 0.01188205,
      "auxiliary_loss_mlp": 0.01029252,
      "balance_loss_clip": 1.05724001,
      "balance_loss_mlp": 1.02163386,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 3.006999797985046,
      "language_loss": 0.83430827,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85648286,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.670060157775879
    },
    {
      "auxiliary_loss_clip": 0.01185343,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.05558634,
      "balance_loss_mlp": 1.01800418,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.595321513612098,
      "language_loss": 0.75646782,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77858078,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.6926474571228027
    },
    {
      "auxiliary_loss_clip": 0.01192696,
      "auxiliary_loss_mlp": 0.01031734,
      "balance_loss_clip": 1.05734456,
      "balance_loss_mlp": 1.02307367,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 2.7045879804669983,
      "language_loss": 0.72689372,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.749138,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 3.5154991149902344
    },
    {
      "auxiliary_loss_clip": 0.01166224,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.04912305,
      "balance_loss_mlp": 1.02093649,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 1.7289825351825394,
      "language_loss": 0.81134784,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83330876,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 3.631861448287964
    },
    {
      "auxiliary_loss_clip": 0.01187713,
      "auxiliary_loss_mlp": 0.01035918,
      "balance_loss_clip": 1.05715036,
      "balance_loss_mlp": 1.02734137,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.0292769416002034,
      "language_loss": 0.70816112,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.7303974,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.5773024559020996
    },
    {
      "auxiliary_loss_clip": 0.01195926,
      "auxiliary_loss_mlp": 0.01031144,
      "balance_loss_clip": 1.0578202,
      "balance_loss_mlp": 1.02231085,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.6415294703160694,
      "language_loss": 0.7907362,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.81300688,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6329495906829834
    },
    {
      "auxiliary_loss_clip": 0.01082449,
      "auxiliary_loss_mlp": 0.01001612,
      "balance_loss_clip": 1.01936102,
      "balance_loss_mlp": 0.99989587,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8917717537415519,
      "language_loss": 0.6283989,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64923954,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.213176965713501
    },
    {
      "auxiliary_loss_clip": 0.01193447,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.05764234,
      "balance_loss_mlp": 1.01784897,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.1765875248080873,
      "language_loss": 0.74351764,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76571524,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.54687762260437
    },
    {
      "auxiliary_loss_clip": 0.01191658,
      "auxiliary_loss_mlp": 0.01028192,
      "balance_loss_clip": 1.05386519,
      "balance_loss_mlp": 1.01997256,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.776375875278715,
      "language_loss": 0.76501864,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78721714,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 3.589787721633911
    },
    {
      "auxiliary_loss_clip": 0.01177447,
      "auxiliary_loss_mlp": 0.01024477,
      "balance_loss_clip": 1.05159104,
      "balance_loss_mlp": 1.01540554,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 2.044321518503995,
      "language_loss": 0.73737359,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75939286,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.63883638381958
    },
    {
      "auxiliary_loss_clip": 0.011848,
      "auxiliary_loss_mlp": 0.01028006,
      "balance_loss_clip": 1.05346751,
      "balance_loss_mlp": 1.01932144,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.4356832668616173,
      "language_loss": 0.76666862,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78879672,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.6267616748809814
    },
    {
      "auxiliary_loss_clip": 0.01184766,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.05509686,
      "balance_loss_mlp": 1.0247823,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.8596018454656909,
      "language_loss": 0.73848933,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76066732,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.6454720497131348
    },
    {
      "auxiliary_loss_clip": 0.01082971,
      "auxiliary_loss_mlp": 0.01001588,
      "balance_loss_clip": 1.01898575,
      "balance_loss_mlp": 1.00006235,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.8518528129296696,
      "language_loss": 0.62815094,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64899647,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.269524574279785
    },
    {
      "auxiliary_loss_clip": 0.01186706,
      "auxiliary_loss_mlp": 0.01030988,
      "balance_loss_clip": 1.05633426,
      "balance_loss_mlp": 1.02264905,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 1.9145451498967991,
      "language_loss": 0.78666496,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80884188,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 2.613227605819702
    },
    {
      "auxiliary_loss_clip": 0.01170885,
      "auxiliary_loss_mlp": 0.01038817,
      "balance_loss_clip": 1.0547241,
      "balance_loss_mlp": 1.02990651,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.8208870123833312,
      "language_loss": 0.79160571,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.8137027,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.670419216156006
    },
    {
      "auxiliary_loss_clip": 0.01159579,
      "auxiliary_loss_mlp": 0.01036116,
      "balance_loss_clip": 1.05055177,
      "balance_loss_mlp": 1.02716386,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 1.8137947718395988,
      "language_loss": 0.83348525,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85544217,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.7069168090820312
    },
    {
      "auxiliary_loss_clip": 0.01072961,
      "auxiliary_loss_mlp": 0.01003296,
      "balance_loss_clip": 1.01867938,
      "balance_loss_mlp": 1.00186527,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7243316688476134,
      "language_loss": 0.53018755,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55095011,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.286557674407959
    },
    {
      "auxiliary_loss_clip": 0.0116834,
      "auxiliary_loss_mlp": 0.01030585,
      "balance_loss_clip": 1.05205619,
      "balance_loss_mlp": 1.0225023,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 2.7096528298005147,
      "language_loss": 0.82630831,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.8482976,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.7914559841156006
    },
    {
      "auxiliary_loss_clip": 0.01040104,
      "auxiliary_loss_mlp": 0.01002263,
      "balance_loss_clip": 1.0138154,
      "balance_loss_mlp": 1.0008744,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.9300027582828957,
      "language_loss": 0.59404165,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.6144653,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.413973093032837
    },
    {
      "auxiliary_loss_clip": 0.0115805,
      "auxiliary_loss_mlp": 0.010263,
      "balance_loss_clip": 1.04604244,
      "balance_loss_mlp": 1.01827717,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.614519408834484,
      "language_loss": 0.78894854,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.81079203,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.845179796218872
    },
    {
      "auxiliary_loss_clip": 0.01166481,
      "auxiliary_loss_mlp": 0.0103018,
      "balance_loss_clip": 1.0500294,
      "balance_loss_mlp": 1.02103043,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 1.87342447853845,
      "language_loss": 0.84151965,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86348629,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.6513681411743164
    },
    {
      "auxiliary_loss_clip": 0.01062746,
      "auxiliary_loss_mlp": 0.01008748,
      "balance_loss_clip": 1.01797521,
      "balance_loss_mlp": 1.00726366,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7314987176639816,
      "language_loss": 0.53874731,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.55946231,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.1726057529449463
    },
    {
      "auxiliary_loss_clip": 0.01173565,
      "auxiliary_loss_mlp": 0.01028398,
      "balance_loss_clip": 1.05460846,
      "balance_loss_mlp": 1.01946962,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 3.1755282914377183,
      "language_loss": 0.78679055,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.80881017,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.6693596839904785
    },
    {
      "auxiliary_loss_clip": 0.01167503,
      "auxiliary_loss_mlp": 0.00902038,
      "balance_loss_clip": 1.05010748,
      "balance_loss_mlp": 1.00108421,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.536652126985226,
      "language_loss": 0.77973413,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80042952,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.734142780303955
    },
    {
      "auxiliary_loss_clip": 0.01089422,
      "auxiliary_loss_mlp": 0.01001649,
      "balance_loss_clip": 1.01781011,
      "balance_loss_mlp": 1.00025427,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.810567002509849,
      "language_loss": 0.59247661,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61338735,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 4.445037364959717
    },
    {
      "auxiliary_loss_clip": 0.01193793,
      "auxiliary_loss_mlp": 0.01028177,
      "balance_loss_clip": 1.05780697,
      "balance_loss_mlp": 1.01946247,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.2346030741648026,
      "language_loss": 0.68244362,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.70466328,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.6031429767608643
    },
    {
      "auxiliary_loss_clip": 0.01188466,
      "auxiliary_loss_mlp": 0.01034872,
      "balance_loss_clip": 1.05740499,
      "balance_loss_mlp": 1.02714097,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.8313515044723616,
      "language_loss": 0.82586694,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84810024,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.657705783843994
    },
    {
      "auxiliary_loss_clip": 0.01149926,
      "auxiliary_loss_mlp": 0.01031296,
      "balance_loss_clip": 1.05090344,
      "balance_loss_mlp": 1.02239108,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.754187780749576,
      "language_loss": 0.76801455,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.78982681,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.7122421264648438
    },
    {
      "auxiliary_loss_clip": 0.01143785,
      "auxiliary_loss_mlp": 0.01023405,
      "balance_loss_clip": 1.04901743,
      "balance_loss_mlp": 1.01484609,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 1.9679186618680735,
      "language_loss": 0.78420848,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80588031,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.7542648315429688
    },
    {
      "auxiliary_loss_clip": 0.01164502,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.05438185,
      "balance_loss_mlp": 1.02267849,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.1340876854319095,
      "language_loss": 0.66421771,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68617463,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 3.7494497299194336
    },
    {
      "auxiliary_loss_clip": 0.01185805,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.05513811,
      "balance_loss_mlp": 1.02462912,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.8904527407684408,
      "language_loss": 0.76483178,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78701979,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 3.5358998775482178
    },
    {
      "auxiliary_loss_clip": 0.01176154,
      "auxiliary_loss_mlp": 0.01032289,
      "balance_loss_clip": 1.05516505,
      "balance_loss_mlp": 1.02388525,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 1.7771571630589498,
      "language_loss": 0.80115902,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.8232435,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.703540325164795
    },
    {
      "auxiliary_loss_clip": 0.01164376,
      "auxiliary_loss_mlp": 0.01035547,
      "balance_loss_clip": 1.05237842,
      "balance_loss_mlp": 1.02646971,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 2.32107102380684,
      "language_loss": 0.76134086,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78334004,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.685906410217285
    },
    {
      "auxiliary_loss_clip": 0.01184228,
      "auxiliary_loss_mlp": 0.0102397,
      "balance_loss_clip": 1.05462134,
      "balance_loss_mlp": 1.01551819,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.239244913802845,
      "language_loss": 0.74513531,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76721728,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.708313226699829
    },
    {
      "auxiliary_loss_clip": 0.01162995,
      "auxiliary_loss_mlp": 0.00902328,
      "balance_loss_clip": 1.04826188,
      "balance_loss_mlp": 1.00097251,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.8657169194593117,
      "language_loss": 0.80251551,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82316875,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 3.6593031883239746
    },
    {
      "auxiliary_loss_clip": 0.01177585,
      "auxiliary_loss_mlp": 0.0102816,
      "balance_loss_clip": 1.05773079,
      "balance_loss_mlp": 1.02015519,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.7964929841142603,
      "language_loss": 0.74368709,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76574457,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.63668155670166
    },
    {
      "auxiliary_loss_clip": 0.01188254,
      "auxiliary_loss_mlp": 0.01031682,
      "balance_loss_clip": 1.05901003,
      "balance_loss_mlp": 1.02296746,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.7712653501749147,
      "language_loss": 0.83951879,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86171812,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.682027816772461
    },
    {
      "auxiliary_loss_clip": 0.01163511,
      "auxiliary_loss_mlp": 0.01036568,
      "balance_loss_clip": 1.05163753,
      "balance_loss_mlp": 1.02770519,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.746551307818847,
      "language_loss": 0.7171675,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73916829,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.767319440841675
    },
    {
      "auxiliary_loss_clip": 0.01179666,
      "auxiliary_loss_mlp": 0.0103524,
      "balance_loss_clip": 1.05366802,
      "balance_loss_mlp": 1.02652621,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 1.8319014921676804,
      "language_loss": 0.75276577,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77491486,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.777646541595459
    },
    {
      "auxiliary_loss_clip": 0.01174625,
      "auxiliary_loss_mlp": 0.01029403,
      "balance_loss_clip": 1.05294359,
      "balance_loss_mlp": 1.02000904,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.02541714057525,
      "language_loss": 0.78767103,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.80971128,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.7273905277252197
    },
    {
      "auxiliary_loss_clip": 0.01143588,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.04668438,
      "balance_loss_mlp": 1.02127492,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 2.0788364727681694,
      "language_loss": 0.76279485,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78453732,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.818354368209839
    },
    {
      "auxiliary_loss_clip": 0.01184393,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.05798936,
      "balance_loss_mlp": 1.02101088,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 4.3101666591586225,
      "language_loss": 0.78342223,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.80555868,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.695425510406494
    },
    {
      "auxiliary_loss_clip": 0.01150949,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 1.05134249,
      "balance_loss_mlp": 1.02275729,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 4.009352944646736,
      "language_loss": 0.806005,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82782638,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.776249885559082
    },
    {
      "auxiliary_loss_clip": 0.01169396,
      "auxiliary_loss_mlp": 0.01025494,
      "balance_loss_clip": 1.05199575,
      "balance_loss_mlp": 1.01762068,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.89069760899885,
      "language_loss": 0.79766321,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81961215,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.7281267642974854
    },
    {
      "auxiliary_loss_clip": 0.01199362,
      "auxiliary_loss_mlp": 0.01033026,
      "balance_loss_clip": 1.06033421,
      "balance_loss_mlp": 1.02387071,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.051514477232705,
      "language_loss": 0.77177131,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79409528,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.562903881072998
    },
    {
      "auxiliary_loss_clip": 0.01171581,
      "auxiliary_loss_mlp": 0.01032696,
      "balance_loss_clip": 1.05269408,
      "balance_loss_mlp": 1.02304626,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 2.9131685266711544,
      "language_loss": 0.79493845,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.8169812,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.7120399475097656
    },
    {
      "auxiliary_loss_clip": 0.01175985,
      "auxiliary_loss_mlp": 0.01037088,
      "balance_loss_clip": 1.05604792,
      "balance_loss_mlp": 1.02806973,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.136110907251353,
      "language_loss": 0.66470206,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68683279,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.677212715148926
    },
    {
      "auxiliary_loss_clip": 0.01194188,
      "auxiliary_loss_mlp": 0.00901911,
      "balance_loss_clip": 1.05783451,
      "balance_loss_mlp": 1.00096309,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.6754892997031312,
      "language_loss": 0.79819739,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81915838,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.6467764377593994
    },
    {
      "auxiliary_loss_clip": 0.01182496,
      "auxiliary_loss_mlp": 0.0103618,
      "balance_loss_clip": 1.05870724,
      "balance_loss_mlp": 1.02741778,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.502807376526822,
      "language_loss": 0.84384918,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86603594,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.5994791984558105
    },
    {
      "auxiliary_loss_clip": 0.01196862,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 1.05734348,
      "balance_loss_mlp": 1.02209115,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 4.749264146038459,
      "language_loss": 0.73485184,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75713634,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.605480194091797
    },
    {
      "auxiliary_loss_clip": 0.01183065,
      "auxiliary_loss_mlp": 0.01029186,
      "balance_loss_clip": 1.05505562,
      "balance_loss_mlp": 1.02065623,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 2.0712921358689633,
      "language_loss": 0.80867219,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.83079469,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.599940776824951
    },
    {
      "auxiliary_loss_clip": 0.01179092,
      "auxiliary_loss_mlp": 0.01028532,
      "balance_loss_clip": 1.05704284,
      "balance_loss_mlp": 1.02027118,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 4.307635661537036,
      "language_loss": 0.75126886,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.77334511,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 3.8540539741516113
    },
    {
      "auxiliary_loss_clip": 0.01193614,
      "auxiliary_loss_mlp": 0.01027491,
      "balance_loss_clip": 1.05721569,
      "balance_loss_mlp": 1.01843095,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 39.77445943320188,
      "language_loss": 0.6544891,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67670012,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.672377109527588
    },
    {
      "auxiliary_loss_clip": 0.01152674,
      "auxiliary_loss_mlp": 0.01030559,
      "balance_loss_clip": 1.05024934,
      "balance_loss_mlp": 1.02083778,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 1.9502238267761396,
      "language_loss": 0.79413581,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81596816,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.765990972518921
    },
    {
      "auxiliary_loss_clip": 0.01197825,
      "auxiliary_loss_mlp": 0.01029136,
      "balance_loss_clip": 1.06022203,
      "balance_loss_mlp": 1.02028501,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 2.389976309110953,
      "language_loss": 0.79454637,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.81681597,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.639331817626953
    },
    {
      "auxiliary_loss_clip": 0.01188361,
      "auxiliary_loss_mlp": 0.01030715,
      "balance_loss_clip": 1.05901492,
      "balance_loss_mlp": 1.02244782,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.80878987606295,
      "language_loss": 0.80682552,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82901627,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.619673252105713
    },
    {
      "auxiliary_loss_clip": 0.01185732,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05761385,
      "balance_loss_mlp": 1.02175331,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 2.087169557159328,
      "language_loss": 0.84847116,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.87062919,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 4.448488473892212
    },
    {
      "auxiliary_loss_clip": 0.01189631,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.05907202,
      "balance_loss_mlp": 1.02061391,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 2.069093851472296,
      "language_loss": 0.79054272,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81272304,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.623626232147217
    },
    {
      "auxiliary_loss_clip": 0.01172598,
      "auxiliary_loss_mlp": 0.01030749,
      "balance_loss_clip": 1.05478144,
      "balance_loss_mlp": 1.02160597,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 5.280432477917545,
      "language_loss": 0.72170925,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74374276,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.6877899169921875
    },
    {
      "auxiliary_loss_clip": 0.01179669,
      "auxiliary_loss_mlp": 0.01033347,
      "balance_loss_clip": 1.05759239,
      "balance_loss_mlp": 1.02391171,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.41488197356006,
      "language_loss": 0.64394778,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66607797,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.7198126316070557
    },
    {
      "auxiliary_loss_clip": 0.01175493,
      "auxiliary_loss_mlp": 0.01031156,
      "balance_loss_clip": 1.05579352,
      "balance_loss_mlp": 1.02287102,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 3.798694994869183,
      "language_loss": 0.774656,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79672253,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 3.6659045219421387
    },
    {
      "auxiliary_loss_clip": 0.01196659,
      "auxiliary_loss_mlp": 0.0102489,
      "balance_loss_clip": 1.05764151,
      "balance_loss_mlp": 1.01655149,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 12.601747811011773,
      "language_loss": 0.70579243,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72800791,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.5628418922424316
    },
    {
      "auxiliary_loss_clip": 0.01163887,
      "auxiliary_loss_mlp": 0.01034801,
      "balance_loss_clip": 1.05548382,
      "balance_loss_mlp": 1.02597344,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.495682514741957,
      "language_loss": 0.78950965,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.8114965,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.745173931121826
    },
    {
      "auxiliary_loss_clip": 0.01184762,
      "auxiliary_loss_mlp": 0.01027088,
      "balance_loss_clip": 1.05886793,
      "balance_loss_mlp": 1.01874948,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.4732350574630764,
      "language_loss": 0.67419797,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69631648,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.6467366218566895
    },
    {
      "auxiliary_loss_clip": 0.01176595,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.05265427,
      "balance_loss_mlp": 1.02082849,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.3218463673703162,
      "language_loss": 0.69032419,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71238583,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.70143985748291
    },
    {
      "auxiliary_loss_clip": 0.01189365,
      "auxiliary_loss_mlp": 0.01032889,
      "balance_loss_clip": 1.05625308,
      "balance_loss_mlp": 1.02424622,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 2.17163057557792,
      "language_loss": 0.79734117,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81956375,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.651776075363159
    },
    {
      "auxiliary_loss_clip": 0.01184214,
      "auxiliary_loss_mlp": 0.01029338,
      "balance_loss_clip": 1.0544951,
      "balance_loss_mlp": 1.0205884,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.787306535625832,
      "language_loss": 0.73872346,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76085901,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 2.7849950790405273
    },
    {
      "auxiliary_loss_clip": 0.01168731,
      "auxiliary_loss_mlp": 0.01032651,
      "balance_loss_clip": 1.05547094,
      "balance_loss_mlp": 1.02428889,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 2.6278903347779266,
      "language_loss": 0.89702439,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91903818,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.71364164352417
    },
    {
      "auxiliary_loss_clip": 0.01184038,
      "auxiliary_loss_mlp": 0.01026404,
      "balance_loss_clip": 1.05774999,
      "balance_loss_mlp": 1.01834607,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.8122186007469057,
      "language_loss": 0.79005122,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.8121556,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.6232125759124756
    },
    {
      "auxiliary_loss_clip": 0.01166796,
      "auxiliary_loss_mlp": 0.0102701,
      "balance_loss_clip": 1.05347323,
      "balance_loss_mlp": 1.01769996,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 2.1181604131212683,
      "language_loss": 0.809021,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83095908,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.67276930809021
    },
    {
      "auxiliary_loss_clip": 0.01173757,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.05560601,
      "balance_loss_mlp": 1.02289963,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.807412115283736,
      "language_loss": 0.74117625,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.7632277,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.682260751724243
    },
    {
      "auxiliary_loss_clip": 0.01195078,
      "auxiliary_loss_mlp": 0.01033676,
      "balance_loss_clip": 1.05769515,
      "balance_loss_mlp": 1.0245924,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.2196185535855752,
      "language_loss": 0.7660718,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.78835934,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.5617854595184326
    },
    {
      "auxiliary_loss_clip": 0.01190094,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.05680442,
      "balance_loss_mlp": 1.02295148,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.685155170237479,
      "language_loss": 0.823704,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84592247,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.6145565509796143
    },
    {
      "auxiliary_loss_clip": 0.0119052,
      "auxiliary_loss_mlp": 0.01031342,
      "balance_loss_clip": 1.05943084,
      "balance_loss_mlp": 1.02274132,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.6665670179924565,
      "language_loss": 0.80935389,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.83157253,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.607668161392212
    },
    {
      "auxiliary_loss_clip": 0.01186323,
      "auxiliary_loss_mlp": 0.01030171,
      "balance_loss_clip": 1.05530334,
      "balance_loss_mlp": 1.02233326,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 2.145224610457806,
      "language_loss": 0.75409013,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77625507,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.672624111175537
    },
    {
      "auxiliary_loss_clip": 0.0117264,
      "auxiliary_loss_mlp": 0.01035894,
      "balance_loss_clip": 1.05569863,
      "balance_loss_mlp": 1.02734709,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.482193216912782,
      "language_loss": 0.75305206,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77513736,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.762392044067383
    },
    {
      "auxiliary_loss_clip": 0.01089109,
      "auxiliary_loss_mlp": 0.00891599,
      "balance_loss_clip": 1.02387285,
      "balance_loss_mlp": 0.99984962,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.8317340876212129,
      "language_loss": 0.6278013,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.6476084,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.2565319538116455
    },
    {
      "auxiliary_loss_clip": 0.01155693,
      "auxiliary_loss_mlp": 0.01031883,
      "balance_loss_clip": 1.04533064,
      "balance_loss_mlp": 1.0234015,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 2.6968942634714894,
      "language_loss": 0.79076052,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81263632,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.756652593612671
    },
    {
      "auxiliary_loss_clip": 0.01195815,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.0578928,
      "balance_loss_mlp": 1.02105081,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 1.897628909345927,
      "language_loss": 0.78969181,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81194544,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 3.592733860015869
    },
    {
      "auxiliary_loss_clip": 0.01170998,
      "auxiliary_loss_mlp": 0.01037746,
      "balance_loss_clip": 1.05263197,
      "balance_loss_mlp": 1.02848899,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 2.9707775179522655,
      "language_loss": 0.6365875,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.65867496,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.709592580795288
    },
    {
      "auxiliary_loss_clip": 0.01184413,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.05453134,
      "balance_loss_mlp": 1.01872778,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 3.0602777301499624,
      "language_loss": 0.85780597,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87991321,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.6293482780456543
    },
    {
      "auxiliary_loss_clip": 0.01197792,
      "auxiliary_loss_mlp": 0.01022361,
      "balance_loss_clip": 1.05908823,
      "balance_loss_mlp": 1.01448178,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 2.11313775579603,
      "language_loss": 0.84550041,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86770195,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.6232593059539795
    },
    {
      "auxiliary_loss_clip": 0.01068354,
      "auxiliary_loss_mlp": 0.01008966,
      "balance_loss_clip": 1.02113652,
      "balance_loss_mlp": 1.00726128,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.7908225402231284,
      "language_loss": 0.57435668,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59512991,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 4.196667194366455
    },
    {
      "auxiliary_loss_clip": 0.01200946,
      "auxiliary_loss_mlp": 0.01034262,
      "balance_loss_clip": 1.06215167,
      "balance_loss_mlp": 1.02557158,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.061556658592786,
      "language_loss": 0.81644475,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.83879685,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 3.6508657932281494
    },
    {
      "auxiliary_loss_clip": 0.01171591,
      "auxiliary_loss_mlp": 0.01033338,
      "balance_loss_clip": 1.05713975,
      "balance_loss_mlp": 1.02536893,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.7281225406723788,
      "language_loss": 0.73760664,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75965595,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.699455976486206
    },
    {
      "auxiliary_loss_clip": 0.01191511,
      "auxiliary_loss_mlp": 0.01028452,
      "balance_loss_clip": 1.05807662,
      "balance_loss_mlp": 1.02000594,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.6898768054366453,
      "language_loss": 0.6555289,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.67772853,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.709573268890381
    },
    {
      "auxiliary_loss_clip": 0.01186658,
      "auxiliary_loss_mlp": 0.01032333,
      "balance_loss_clip": 1.05956638,
      "balance_loss_mlp": 1.0246079,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.8266960171351196,
      "language_loss": 0.82240266,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84459251,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.681267023086548
    },
    {
      "auxiliary_loss_clip": 0.01170553,
      "auxiliary_loss_mlp": 0.01030932,
      "balance_loss_clip": 1.05096161,
      "balance_loss_mlp": 1.0223372,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 1.92995232915793,
      "language_loss": 0.82371098,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84572583,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 3.698615789413452
    },
    {
      "auxiliary_loss_clip": 0.01165617,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.05250251,
      "balance_loss_mlp": 1.01807451,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.5542446448814808,
      "language_loss": 0.68227077,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70418882,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.771996259689331
    },
    {
      "auxiliary_loss_clip": 0.01174821,
      "auxiliary_loss_mlp": 0.01030002,
      "balance_loss_clip": 1.05619109,
      "balance_loss_mlp": 1.02216363,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 2.0572319945211954,
      "language_loss": 0.74591684,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76796508,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.6299524307250977
    },
    {
      "auxiliary_loss_clip": 0.01162962,
      "auxiliary_loss_mlp": 0.01027416,
      "balance_loss_clip": 1.05139112,
      "balance_loss_mlp": 1.01948309,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.6923071501089684,
      "language_loss": 0.80858696,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.83049077,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.708080768585205
    },
    {
      "auxiliary_loss_clip": 0.01185806,
      "auxiliary_loss_mlp": 0.00901893,
      "balance_loss_clip": 1.05861783,
      "balance_loss_mlp": 1.00069022,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 3.4767289886418458,
      "language_loss": 0.90505493,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.92593193,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.66752028465271
    },
    {
      "auxiliary_loss_clip": 0.01170327,
      "auxiliary_loss_mlp": 0.01032348,
      "balance_loss_clip": 1.0571425,
      "balance_loss_mlp": 1.02421808,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.927447410602192,
      "language_loss": 0.76639628,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78842306,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.6683201789855957
    },
    {
      "auxiliary_loss_clip": 0.01096789,
      "auxiliary_loss_mlp": 0.01005415,
      "balance_loss_clip": 1.0249691,
      "balance_loss_mlp": 1.00415695,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7294607933976212,
      "language_loss": 0.58728516,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.60830724,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.237412929534912
    },
    {
      "auxiliary_loss_clip": 0.01193132,
      "auxiliary_loss_mlp": 0.01027743,
      "balance_loss_clip": 1.05915248,
      "balance_loss_mlp": 1.01990521,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.7833135701894363,
      "language_loss": 0.70038658,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72259533,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.586951971054077
    },
    {
      "auxiliary_loss_clip": 0.01178832,
      "auxiliary_loss_mlp": 0.01030349,
      "balance_loss_clip": 1.05197775,
      "balance_loss_mlp": 1.02134836,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.819455158446813,
      "language_loss": 0.83586919,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.857961,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.5863189697265625
    },
    {
      "auxiliary_loss_clip": 0.01172261,
      "auxiliary_loss_mlp": 0.00901572,
      "balance_loss_clip": 1.05244505,
      "balance_loss_mlp": 1.00063729,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.9751287053014803,
      "language_loss": 0.73311651,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.75385487,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.713970422744751
    },
    {
      "auxiliary_loss_clip": 0.01182347,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.05830419,
      "balance_loss_mlp": 1.01722884,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.734874459093155,
      "language_loss": 0.81040913,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83248389,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.6851937770843506
    },
    {
      "auxiliary_loss_clip": 0.0117982,
      "auxiliary_loss_mlp": 0.01023156,
      "balance_loss_clip": 1.05554938,
      "balance_loss_mlp": 1.01597965,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.7234402915786105,
      "language_loss": 0.8309294,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85295916,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.715237617492676
    },
    {
      "auxiliary_loss_clip": 0.01177094,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.0558573,
      "balance_loss_mlp": 1.01926124,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 1.8078438389740057,
      "language_loss": 0.8200556,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84210205,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.714142084121704
    },
    {
      "auxiliary_loss_clip": 0.01133097,
      "auxiliary_loss_mlp": 0.01035221,
      "balance_loss_clip": 1.05048704,
      "balance_loss_mlp": 1.02636969,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 3.793137801656045,
      "language_loss": 0.73422688,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75591004,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.7190756797790527
    },
    {
      "auxiliary_loss_clip": 0.01186056,
      "auxiliary_loss_mlp": 0.00900155,
      "balance_loss_clip": 1.05766618,
      "balance_loss_mlp": 1.00066304,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.028859956893524,
      "language_loss": 0.8212167,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.84207886,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.6481571197509766
    },
    {
      "auxiliary_loss_clip": 0.01176495,
      "auxiliary_loss_mlp": 0.0103373,
      "balance_loss_clip": 1.0517087,
      "balance_loss_mlp": 1.02549863,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 2.0127722957988676,
      "language_loss": 0.79818159,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.82028383,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.6559560298919678
    },
    {
      "auxiliary_loss_clip": 0.01131428,
      "auxiliary_loss_mlp": 0.01031345,
      "balance_loss_clip": 1.05190682,
      "balance_loss_mlp": 1.0232985,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.3405952881569134,
      "language_loss": 0.84166288,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86329061,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.9684741497039795
    },
    {
      "auxiliary_loss_clip": 0.01087968,
      "auxiliary_loss_mlp": 0.01001688,
      "balance_loss_clip": 1.02445555,
      "balance_loss_mlp": 1.00041795,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9801235376106617,
      "language_loss": 0.72127008,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74216664,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.467947244644165
    },
    {
      "auxiliary_loss_clip": 0.01164313,
      "auxiliary_loss_mlp": 0.01025488,
      "balance_loss_clip": 1.05603492,
      "balance_loss_mlp": 1.01871443,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 1.8661564501920058,
      "language_loss": 0.81996894,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.84186697,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 3.7078824043273926
    },
    {
      "auxiliary_loss_clip": 0.01089607,
      "auxiliary_loss_mlp": 0.01008114,
      "balance_loss_clip": 1.04003322,
      "balance_loss_mlp": 1.00674868,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7591994756436911,
      "language_loss": 0.54910672,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.57008392,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.1935205459594727
    },
    {
      "auxiliary_loss_clip": 0.01185324,
      "auxiliary_loss_mlp": 0.01029765,
      "balance_loss_clip": 1.0561552,
      "balance_loss_mlp": 1.02153993,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.148652272373337,
      "language_loss": 0.7696681,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79181898,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.649852752685547
    },
    {
      "auxiliary_loss_clip": 0.01181754,
      "auxiliary_loss_mlp": 0.01026824,
      "balance_loss_clip": 1.05772924,
      "balance_loss_mlp": 1.01892078,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 1.9996631532218199,
      "language_loss": 0.88479334,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90687913,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.681722402572632
    },
    {
      "auxiliary_loss_clip": 0.01199346,
      "auxiliary_loss_mlp": 0.01031214,
      "balance_loss_clip": 1.06088233,
      "balance_loss_mlp": 1.02283943,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 3.481768192297674,
      "language_loss": 0.93776578,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.96007133,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 3.6081883907318115
    },
    {
      "auxiliary_loss_clip": 0.01185958,
      "auxiliary_loss_mlp": 0.01037345,
      "balance_loss_clip": 1.05756044,
      "balance_loss_mlp": 1.03001368,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.7973450791061314,
      "language_loss": 0.73310041,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75533342,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 3.5557820796966553
    },
    {
      "auxiliary_loss_clip": 0.0116032,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04927349,
      "balance_loss_mlp": 1.01908422,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 1.868274361624464,
      "language_loss": 0.85327971,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87515861,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.7062647342681885
    },
    {
      "auxiliary_loss_clip": 0.01170803,
      "auxiliary_loss_mlp": 0.00901217,
      "balance_loss_clip": 1.0567205,
      "balance_loss_mlp": 1.00070524,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.7721648937999825,
      "language_loss": 0.84541547,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86613572,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.674332618713379
    },
    {
      "auxiliary_loss_clip": 0.01157112,
      "auxiliary_loss_mlp": 0.01033061,
      "balance_loss_clip": 1.04942536,
      "balance_loss_mlp": 1.02400112,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.2037615670663486,
      "language_loss": 0.79779077,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.81969249,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.7484662532806396
    },
    {
      "auxiliary_loss_clip": 0.01158872,
      "auxiliary_loss_mlp": 0.01030372,
      "balance_loss_clip": 1.0543015,
      "balance_loss_mlp": 1.02223873,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 1.7703500855949315,
      "language_loss": 0.89785826,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91975069,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 3.6581811904907227
    },
    {
      "auxiliary_loss_clip": 0.01196851,
      "auxiliary_loss_mlp": 0.01028154,
      "balance_loss_clip": 1.05834854,
      "balance_loss_mlp": 1.01973224,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.521278702807373,
      "language_loss": 0.80882537,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83107543,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.559281349182129
    },
    {
      "auxiliary_loss_clip": 0.0116666,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.05586624,
      "balance_loss_mlp": 1.01954389,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 1.7234550924278442,
      "language_loss": 0.74575627,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76770222,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.687959671020508
    },
    {
      "auxiliary_loss_clip": 0.01189277,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.06076479,
      "balance_loss_mlp": 1.02111697,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 1.8787967146297322,
      "language_loss": 0.87029272,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89247417,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.6095080375671387
    },
    {
      "auxiliary_loss_clip": 0.01177507,
      "auxiliary_loss_mlp": 0.01030914,
      "balance_loss_clip": 1.05764437,
      "balance_loss_mlp": 1.02292132,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.6761634844538547,
      "language_loss": 0.87880826,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.9008925,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.7194156646728516
    },
    {
      "auxiliary_loss_clip": 0.01176738,
      "auxiliary_loss_mlp": 0.01029174,
      "balance_loss_clip": 1.05431688,
      "balance_loss_mlp": 1.02065647,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 2.0794642553200555,
      "language_loss": 0.85198033,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87403947,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.6237099170684814
    },
    {
      "auxiliary_loss_clip": 0.01153706,
      "auxiliary_loss_mlp": 0.01024327,
      "balance_loss_clip": 1.05198741,
      "balance_loss_mlp": 1.01607144,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.296732821069544,
      "language_loss": 0.74699897,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76877928,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.7581536769866943
    },
    {
      "auxiliary_loss_clip": 0.01082645,
      "auxiliary_loss_mlp": 0.01008685,
      "balance_loss_clip": 1.02133036,
      "balance_loss_mlp": 1.00731397,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7912324748087398,
      "language_loss": 0.6204136,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.6413269,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.2136459350585938
    },
    {
      "auxiliary_loss_clip": 0.01192964,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.05788612,
      "balance_loss_mlp": 1.02015865,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 1.9611461560889747,
      "language_loss": 0.80748272,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.82968813,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.584223747253418
    },
    {
      "auxiliary_loss_clip": 0.01185457,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.05752814,
      "balance_loss_mlp": 1.02045512,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 2.0148192541367203,
      "language_loss": 0.75984496,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78198212,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.6228554248809814
    },
    {
      "auxiliary_loss_clip": 0.01174165,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.05991757,
      "balance_loss_mlp": 1.02081895,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 1.8985234019391886,
      "language_loss": 0.76325548,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78528965,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.6203763484954834
    },
    {
      "auxiliary_loss_clip": 0.0118,
      "auxiliary_loss_mlp": 0.01032685,
      "balance_loss_clip": 1.05636835,
      "balance_loss_mlp": 1.02440548,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 2.176017027056813,
      "language_loss": 0.85038209,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.872509,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.6464197635650635
    },
    {
      "auxiliary_loss_clip": 0.01193719,
      "auxiliary_loss_mlp": 0.01026936,
      "balance_loss_clip": 1.0621295,
      "balance_loss_mlp": 1.0190289,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.986186411110815,
      "language_loss": 0.831779,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85398549,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.618041515350342
    },
    {
      "auxiliary_loss_clip": 0.01184051,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.05742061,
      "balance_loss_mlp": 1.02164567,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.4305792020099117,
      "language_loss": 0.76866579,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79080164,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.645700454711914
    },
    {
      "auxiliary_loss_clip": 0.01196879,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.06028199,
      "balance_loss_mlp": 1.0178566,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 7.79320348663781,
      "language_loss": 0.86964548,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.89187419,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.582697868347168
    },
    {
      "auxiliary_loss_clip": 0.01166162,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.05411804,
      "balance_loss_mlp": 1.02232075,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.5031598755956064,
      "language_loss": 0.66677773,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.68874323,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.6571426391601562
    },
    {
      "auxiliary_loss_clip": 0.01184205,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 1.05650628,
      "balance_loss_mlp": 1.0232538,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.2333679795282113,
      "language_loss": 0.79226959,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.81442767,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.6660759449005127
    },
    {
      "auxiliary_loss_clip": 0.01185578,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.06001854,
      "balance_loss_mlp": 1.02240956,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.786104767565531,
      "language_loss": 0.84618533,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.86834586,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 3.6271817684173584
    },
    {
      "auxiliary_loss_clip": 0.01146719,
      "auxiliary_loss_mlp": 0.01030808,
      "balance_loss_clip": 1.0481956,
      "balance_loss_mlp": 1.02198637,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.352912050457232,
      "language_loss": 0.86898851,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.89076382,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.822895050048828
    },
    {
      "auxiliary_loss_clip": 0.01178226,
      "auxiliary_loss_mlp": 0.01031555,
      "balance_loss_clip": 1.05853772,
      "balance_loss_mlp": 1.02324581,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.4317862450245273,
      "language_loss": 0.7644242,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78652191,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.639885663986206
    },
    {
      "auxiliary_loss_clip": 0.01183881,
      "auxiliary_loss_mlp": 0.01034678,
      "balance_loss_clip": 1.0574801,
      "balance_loss_mlp": 1.02656579,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 2.0680212104909974,
      "language_loss": 0.89054108,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91272676,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.746725559234619
    },
    {
      "auxiliary_loss_clip": 0.01189886,
      "auxiliary_loss_mlp": 0.0103584,
      "balance_loss_clip": 1.05727363,
      "balance_loss_mlp": 1.02729273,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 2.0339387916708356,
      "language_loss": 0.80665338,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82891065,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 3.549710512161255
    },
    {
      "auxiliary_loss_clip": 0.01172528,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 1.05880404,
      "balance_loss_mlp": 1.02118468,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.573840322400691,
      "language_loss": 0.71378732,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73581254,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.779233455657959
    },
    {
      "auxiliary_loss_clip": 0.01185976,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05710006,
      "balance_loss_mlp": 1.02100754,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.831991587092556,
      "language_loss": 0.76507843,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.7872346,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 3.635028839111328
    },
    {
      "auxiliary_loss_clip": 0.01160196,
      "auxiliary_loss_mlp": 0.01028288,
      "balance_loss_clip": 1.05278933,
      "balance_loss_mlp": 1.01981819,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.226714391480001,
      "language_loss": 0.82395691,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84584171,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.6988704204559326
    },
    {
      "auxiliary_loss_clip": 0.01164853,
      "auxiliary_loss_mlp": 0.01029291,
      "balance_loss_clip": 1.058846,
      "balance_loss_mlp": 1.02031469,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 1.888990318126176,
      "language_loss": 0.73702157,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75896299,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.702507495880127
    },
    {
      "auxiliary_loss_clip": 0.01176966,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.06234884,
      "balance_loss_mlp": 1.0262183,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.9355483182663535,
      "language_loss": 0.81132621,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.8334437,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 3.5408618450164795
    },
    {
      "auxiliary_loss_clip": 0.01185772,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.06078267,
      "balance_loss_mlp": 1.02227914,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 5.494744207875245,
      "language_loss": 0.85770261,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.87986904,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.600759506225586
    },
    {
      "auxiliary_loss_clip": 0.01164432,
      "auxiliary_loss_mlp": 0.01031206,
      "balance_loss_clip": 1.0515461,
      "balance_loss_mlp": 1.02241492,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.9074468642940432,
      "language_loss": 0.74463594,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76659238,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.653785467147827
    },
    {
      "auxiliary_loss_clip": 0.01186436,
      "auxiliary_loss_mlp": 0.01027613,
      "balance_loss_clip": 1.05758953,
      "balance_loss_mlp": 1.01857138,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 2.613453251157,
      "language_loss": 0.78311908,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80525959,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.6712028980255127
    },
    {
      "auxiliary_loss_clip": 0.01085447,
      "auxiliary_loss_mlp": 0.01003445,
      "balance_loss_clip": 1.02581191,
      "balance_loss_mlp": 1.00184715,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8282918082119762,
      "language_loss": 0.66073489,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.68162382,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.160130500793457
    },
    {
      "auxiliary_loss_clip": 0.01176442,
      "auxiliary_loss_mlp": 0.00901658,
      "balance_loss_clip": 1.05728507,
      "balance_loss_mlp": 1.00085175,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 3.8883356714017037,
      "language_loss": 0.92108011,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.94186103,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.605433702468872
    },
    {
      "auxiliary_loss_clip": 0.01077298,
      "auxiliary_loss_mlp": 0.01004766,
      "balance_loss_clip": 1.03398681,
      "balance_loss_mlp": 1.00318074,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7805997410603011,
      "language_loss": 0.54717112,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56799179,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.065800905227661
    },
    {
      "auxiliary_loss_clip": 0.01182717,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.05586314,
      "balance_loss_mlp": 1.01982403,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 2.016648741915058,
      "language_loss": 0.80999565,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.83210462,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.623878002166748
    },
    {
      "auxiliary_loss_clip": 0.01101252,
      "auxiliary_loss_mlp": 0.01027812,
      "balance_loss_clip": 1.03853118,
      "balance_loss_mlp": 1.01944327,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.971264191737564,
      "language_loss": 0.76388949,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78518009,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 3.082998752593994
    },
    {
      "auxiliary_loss_clip": 0.01194433,
      "auxiliary_loss_mlp": 0.01032119,
      "balance_loss_clip": 1.05876267,
      "balance_loss_mlp": 1.02399468,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.86233864084983,
      "language_loss": 0.69991934,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.7221849,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.7861487865448
    },
    {
      "auxiliary_loss_clip": 0.01159963,
      "auxiliary_loss_mlp": 0.01025874,
      "balance_loss_clip": 1.05880797,
      "balance_loss_mlp": 1.01737463,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 3.3892613787758292,
      "language_loss": 0.66174817,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68360662,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.787726402282715
    },
    {
      "auxiliary_loss_clip": 0.01174903,
      "auxiliary_loss_mlp": 0.01032774,
      "balance_loss_clip": 1.05541205,
      "balance_loss_mlp": 1.02497184,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 2.854836062256039,
      "language_loss": 0.70699394,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.72907066,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.6430346965789795
    },
    {
      "auxiliary_loss_clip": 0.01091183,
      "auxiliary_loss_mlp": 0.01002479,
      "balance_loss_clip": 1.03526354,
      "balance_loss_mlp": 1.00092924,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7753381136492938,
      "language_loss": 0.61683643,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63777304,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.0958924293518066
    },
    {
      "auxiliary_loss_clip": 0.01146504,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.05147707,
      "balance_loss_mlp": 1.01997733,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.366505768395267,
      "language_loss": 0.76802504,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.78977948,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.759094715118408
    },
    {
      "auxiliary_loss_clip": 0.01174529,
      "auxiliary_loss_mlp": 0.01027436,
      "balance_loss_clip": 1.05955994,
      "balance_loss_mlp": 1.01972902,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.7637511543655129,
      "language_loss": 0.84063303,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86265272,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.680177688598633
    },
    {
      "auxiliary_loss_clip": 0.01171643,
      "auxiliary_loss_mlp": 0.01029106,
      "balance_loss_clip": 1.05542278,
      "balance_loss_mlp": 1.02058887,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.8204923354759777,
      "language_loss": 0.76916212,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79116958,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.6501705646514893
    },
    {
      "auxiliary_loss_clip": 0.01177211,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05603802,
      "balance_loss_mlp": 1.02290297,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.096384163286643,
      "language_loss": 0.67117864,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69327426,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.6453559398651123
    },
    {
      "auxiliary_loss_clip": 0.01188669,
      "auxiliary_loss_mlp": 0.01033038,
      "balance_loss_clip": 1.06017601,
      "balance_loss_mlp": 1.02480054,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.9452906843008633,
      "language_loss": 0.77700186,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79921889,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.609351634979248
    },
    {
      "auxiliary_loss_clip": 0.01181323,
      "auxiliary_loss_mlp": 0.00902089,
      "balance_loss_clip": 1.0560807,
      "balance_loss_mlp": 1.00085878,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 2.239861095670506,
      "language_loss": 0.77612853,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.79696268,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 4.341168403625488
    },
    {
      "auxiliary_loss_clip": 0.01182577,
      "auxiliary_loss_mlp": 0.01032912,
      "balance_loss_clip": 1.05846381,
      "balance_loss_mlp": 1.02390611,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.17913347783043,
      "language_loss": 0.68911946,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71127439,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.816988706588745
    },
    {
      "auxiliary_loss_clip": 0.01177584,
      "auxiliary_loss_mlp": 0.01031967,
      "balance_loss_clip": 1.05530667,
      "balance_loss_mlp": 1.02349138,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.2421748481560533,
      "language_loss": 0.6631313,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.6852268,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.6634280681610107
    },
    {
      "auxiliary_loss_clip": 0.01166652,
      "auxiliary_loss_mlp": 0.010291,
      "balance_loss_clip": 1.04862905,
      "balance_loss_mlp": 1.02007556,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.177396194987605,
      "language_loss": 0.85681432,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87877178,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 3.6186134815216064
    },
    {
      "auxiliary_loss_clip": 0.01178051,
      "auxiliary_loss_mlp": 0.01040244,
      "balance_loss_clip": 1.05950999,
      "balance_loss_mlp": 1.03131461,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 1.6858498728286244,
      "language_loss": 0.85924435,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88142729,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.6463782787323
    },
    {
      "auxiliary_loss_clip": 0.01194335,
      "auxiliary_loss_mlp": 0.0102473,
      "balance_loss_clip": 1.05959713,
      "balance_loss_mlp": 1.01678491,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 1.8924745314242102,
      "language_loss": 0.80850279,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.83069342,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 3.5691568851470947
    },
    {
      "auxiliary_loss_clip": 0.01193648,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 1.06326842,
      "balance_loss_mlp": 1.0217495,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.636217538605745,
      "language_loss": 0.80031908,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.8225559,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.6682350635528564
    },
    {
      "auxiliary_loss_clip": 0.01179971,
      "auxiliary_loss_mlp": 0.0103681,
      "balance_loss_clip": 1.0547334,
      "balance_loss_mlp": 1.02740431,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.647776066475384,
      "language_loss": 0.76840264,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.7905705,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.6215782165527344
    },
    {
      "auxiliary_loss_clip": 0.01195833,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.05902886,
      "balance_loss_mlp": 1.01876915,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.7526763232840006,
      "language_loss": 0.81668562,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.83891416,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.580850124359131
    },
    {
      "auxiliary_loss_clip": 0.01199428,
      "auxiliary_loss_mlp": 0.01032108,
      "balance_loss_clip": 1.05974674,
      "balance_loss_mlp": 1.02260089,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 3.4159215140220236,
      "language_loss": 0.70511699,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72743243,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 3.518047571182251
    },
    {
      "auxiliary_loss_clip": 0.01171135,
      "auxiliary_loss_mlp": 0.01030617,
      "balance_loss_clip": 1.05480587,
      "balance_loss_mlp": 1.02181363,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 2.434491255268036,
      "language_loss": 0.76743817,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78945565,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.652987003326416
    },
    {
      "auxiliary_loss_clip": 0.01179189,
      "auxiliary_loss_mlp": 0.01034691,
      "balance_loss_clip": 1.0577178,
      "balance_loss_mlp": 1.02550006,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.456830677720614,
      "language_loss": 0.71036279,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.73250163,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.69587779045105
    },
    {
      "auxiliary_loss_clip": 0.01096428,
      "auxiliary_loss_mlp": 0.01003454,
      "balance_loss_clip": 1.02826965,
      "balance_loss_mlp": 1.00199401,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6749688620734957,
      "language_loss": 0.57786882,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59886765,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.1693363189697266
    },
    {
      "auxiliary_loss_clip": 0.0117209,
      "auxiliary_loss_mlp": 0.01029692,
      "balance_loss_clip": 1.05381727,
      "balance_loss_mlp": 1.02088249,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.78813094405279,
      "language_loss": 0.75513244,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77715027,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.707650661468506
    },
    {
      "auxiliary_loss_clip": 0.01165039,
      "auxiliary_loss_mlp": 0.0103312,
      "balance_loss_clip": 1.05485177,
      "balance_loss_mlp": 1.02376831,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 2.0193834001224813,
      "language_loss": 0.83527493,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85725653,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.77247953414917
    },
    {
      "auxiliary_loss_clip": 0.01179773,
      "auxiliary_loss_mlp": 0.01032692,
      "balance_loss_clip": 1.05602574,
      "balance_loss_mlp": 1.02418602,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 5.28780332661518,
      "language_loss": 0.78565502,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80777967,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.640657901763916
    },
    {
      "auxiliary_loss_clip": 0.01182273,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.05687046,
      "balance_loss_mlp": 1.02366292,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 1.8761216996906385,
      "language_loss": 0.7978977,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82004333,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.6592023372650146
    },
    {
      "auxiliary_loss_clip": 0.01084947,
      "auxiliary_loss_mlp": 0.01001911,
      "balance_loss_clip": 1.0253818,
      "balance_loss_mlp": 1.00033748,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9430215148074613,
      "language_loss": 0.57092154,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59179014,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.277998924255371
    },
    {
      "auxiliary_loss_clip": 0.01181576,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.05786753,
      "balance_loss_mlp": 1.02684593,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 1.8711833943042078,
      "language_loss": 0.69917238,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.72134471,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.7489161491394043
    },
    {
      "auxiliary_loss_clip": 0.0113234,
      "auxiliary_loss_mlp": 0.01030888,
      "balance_loss_clip": 1.04691863,
      "balance_loss_mlp": 1.02128577,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 1.8858890359504827,
      "language_loss": 0.72934282,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.75097507,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.800797700881958
    },
    {
      "auxiliary_loss_clip": 0.01201397,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.06270337,
      "balance_loss_mlp": 1.02348101,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.7668279084559748,
      "language_loss": 0.78451002,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.8068471,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.5782523155212402
    },
    {
      "auxiliary_loss_clip": 0.01176454,
      "auxiliary_loss_mlp": 0.01032684,
      "balance_loss_clip": 1.05616713,
      "balance_loss_mlp": 1.02402377,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 2.501975632452398,
      "language_loss": 0.78303015,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80512154,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.7905683517456055
    },
    {
      "auxiliary_loss_clip": 0.01072915,
      "auxiliary_loss_mlp": 0.01007636,
      "balance_loss_clip": 1.03126454,
      "balance_loss_mlp": 1.00607395,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.6962399756287786,
      "language_loss": 0.60283351,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62363899,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.314836263656616
    },
    {
      "auxiliary_loss_clip": 0.01096355,
      "auxiliary_loss_mlp": 0.01002753,
      "balance_loss_clip": 1.02353835,
      "balance_loss_mlp": 1.00133419,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8885838385710417,
      "language_loss": 0.57913351,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.6001246,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.1568896770477295
    },
    {
      "auxiliary_loss_clip": 0.01201862,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.06248033,
      "balance_loss_mlp": 1.02351701,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 2.211394932971104,
      "language_loss": 0.76220572,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78454846,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.6864871978759766
    },
    {
      "auxiliary_loss_clip": 0.01090336,
      "auxiliary_loss_mlp": 0.01002007,
      "balance_loss_clip": 1.02156174,
      "balance_loss_mlp": 1.00063658,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.7703454187259908,
      "language_loss": 0.61737674,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63830012,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.187535524368286
    },
    {
      "auxiliary_loss_clip": 0.0118759,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.06106865,
      "balance_loss_mlp": 1.01699543,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.0678360483868463,
      "language_loss": 0.79206157,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81418896,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 3.779073715209961
    },
    {
      "auxiliary_loss_clip": 0.0117034,
      "auxiliary_loss_mlp": 0.009035,
      "balance_loss_clip": 1.05736911,
      "balance_loss_mlp": 1.00100327,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.5371468051431747,
      "language_loss": 0.78835464,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.809093,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.7251548767089844
    },
    {
      "auxiliary_loss_clip": 0.01182604,
      "auxiliary_loss_mlp": 0.01037216,
      "balance_loss_clip": 1.05201221,
      "balance_loss_mlp": 1.02813184,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.607074032286474,
      "language_loss": 0.77972555,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.80192375,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.6779778003692627
    },
    {
      "auxiliary_loss_clip": 0.01165457,
      "auxiliary_loss_mlp": 0.01034904,
      "balance_loss_clip": 1.05400264,
      "balance_loss_mlp": 1.02639234,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 1.9155153793743114,
      "language_loss": 0.74675047,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.76875412,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 3.6499650478363037
    },
    {
      "auxiliary_loss_clip": 0.0120146,
      "auxiliary_loss_mlp": 0.01035062,
      "balance_loss_clip": 1.06267428,
      "balance_loss_mlp": 1.02653837,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 3.4810196710276857,
      "language_loss": 0.78483713,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.8072024,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.6319522857666016
    },
    {
      "auxiliary_loss_clip": 0.01171682,
      "auxiliary_loss_mlp": 0.01030719,
      "balance_loss_clip": 1.05522776,
      "balance_loss_mlp": 1.0221715,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.4954653953539094,
      "language_loss": 0.7289868,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.75101084,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 3.6840078830718994
    },
    {
      "auxiliary_loss_clip": 0.01166109,
      "auxiliary_loss_mlp": 0.01030666,
      "balance_loss_clip": 1.05527329,
      "balance_loss_mlp": 1.02219594,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.8809566998401006,
      "language_loss": 0.80095494,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82292271,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.8187568187713623
    },
    {
      "auxiliary_loss_clip": 0.01168296,
      "auxiliary_loss_mlp": 0.01030301,
      "balance_loss_clip": 1.04902804,
      "balance_loss_mlp": 1.02159309,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.9259530739086375,
      "language_loss": 0.76037335,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78235936,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.6541128158569336
    },
    {
      "auxiliary_loss_clip": 0.01199073,
      "auxiliary_loss_mlp": 0.01042616,
      "balance_loss_clip": 1.05660522,
      "balance_loss_mlp": 1.03358626,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.0079142621901234,
      "language_loss": 0.92120779,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94362462,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.648756980895996
    },
    {
      "auxiliary_loss_clip": 0.01075931,
      "auxiliary_loss_mlp": 0.01002829,
      "balance_loss_clip": 1.01980639,
      "balance_loss_mlp": 1.00144005,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8084498487094002,
      "language_loss": 0.56495798,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58574557,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 4.24988865852356
    },
    {
      "auxiliary_loss_clip": 0.01191106,
      "auxiliary_loss_mlp": 0.01032287,
      "balance_loss_clip": 1.06031239,
      "balance_loss_mlp": 1.02300692,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 2.495392689547672,
      "language_loss": 0.81206363,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.8342976,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.618508815765381
    },
    {
      "auxiliary_loss_clip": 0.01184794,
      "auxiliary_loss_mlp": 0.01030854,
      "balance_loss_clip": 1.05652535,
      "balance_loss_mlp": 1.02193761,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 2.482642019747587,
      "language_loss": 0.8251459,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84730238,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.760206699371338
    },
    {
      "auxiliary_loss_clip": 0.01189632,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.05796981,
      "balance_loss_mlp": 1.01779819,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 2.0960337307148706,
      "language_loss": 0.70821965,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73038995,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.768758535385132
    },
    {
      "auxiliary_loss_clip": 0.01188889,
      "auxiliary_loss_mlp": 0.01033622,
      "balance_loss_clip": 1.05818617,
      "balance_loss_mlp": 1.02519417,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 2.11305690773275,
      "language_loss": 0.88225192,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90447706,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.82615065574646
    },
    {
      "auxiliary_loss_clip": 0.01183856,
      "auxiliary_loss_mlp": 0.01037441,
      "balance_loss_clip": 1.06063116,
      "balance_loss_mlp": 1.02801728,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 1.9364064797100435,
      "language_loss": 0.7942189,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81643188,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.820995807647705
    },
    {
      "auxiliary_loss_clip": 0.01180122,
      "auxiliary_loss_mlp": 0.01029212,
      "balance_loss_clip": 1.05348802,
      "balance_loss_mlp": 1.01978278,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 5.005565119192194,
      "language_loss": 0.81274956,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83484292,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.7262022495269775
    },
    {
      "auxiliary_loss_clip": 0.01171521,
      "auxiliary_loss_mlp": 0.01040279,
      "balance_loss_clip": 1.05448413,
      "balance_loss_mlp": 1.0309155,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.6818084403510405,
      "language_loss": 0.63663471,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65875274,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.6316866874694824
    },
    {
      "auxiliary_loss_clip": 0.01184709,
      "auxiliary_loss_mlp": 0.01030947,
      "balance_loss_clip": 1.05650103,
      "balance_loss_mlp": 1.02203584,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.926922234195051,
      "language_loss": 0.74944794,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77160454,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.690049886703491
    },
    {
      "auxiliary_loss_clip": 0.01197572,
      "auxiliary_loss_mlp": 0.01036659,
      "balance_loss_clip": 1.05805755,
      "balance_loss_mlp": 1.0275631,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 2.2681024459480894,
      "language_loss": 0.85959435,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.88193667,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.6040258407592773
    },
    {
      "auxiliary_loss_clip": 0.01174969,
      "auxiliary_loss_mlp": 0.01033689,
      "balance_loss_clip": 1.05615377,
      "balance_loss_mlp": 1.02512932,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 1.9525765602272622,
      "language_loss": 0.64340311,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66548967,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.7027652263641357
    },
    {
      "auxiliary_loss_clip": 0.01151522,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.04872549,
      "balance_loss_mlp": 1.01968718,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.1676197880853323,
      "language_loss": 0.91790342,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.93970031,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.7612462043762207
    },
    {
      "auxiliary_loss_clip": 0.01165637,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.05233431,
      "balance_loss_mlp": 1.02387452,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.042864343856291,
      "language_loss": 0.75012958,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77212077,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.629672050476074
    },
    {
      "auxiliary_loss_clip": 0.01175884,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.05802417,
      "balance_loss_mlp": 1.0241611,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.8575070005019914,
      "language_loss": 0.65256786,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67464978,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.7320592403411865
    },
    {
      "auxiliary_loss_clip": 0.01179042,
      "auxiliary_loss_mlp": 0.01033017,
      "balance_loss_clip": 1.05707312,
      "balance_loss_mlp": 1.02412951,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 1.9564788006320883,
      "language_loss": 0.68289948,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.70502001,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.598025321960449
    },
    {
      "auxiliary_loss_clip": 0.01177746,
      "auxiliary_loss_mlp": 0.00902806,
      "balance_loss_clip": 1.05543256,
      "balance_loss_mlp": 1.00077677,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 1.920401027406902,
      "language_loss": 0.78309953,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80390513,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.6495327949523926
    },
    {
      "auxiliary_loss_clip": 0.01196445,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.06115055,
      "balance_loss_mlp": 1.02099741,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.8481173324136964,
      "language_loss": 0.72613788,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74839598,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.6048572063446045
    },
    {
      "auxiliary_loss_clip": 0.01190675,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.05609465,
      "balance_loss_mlp": 1.0197618,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.066431711315026,
      "language_loss": 0.8575542,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87974286,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 3.4573099613189697
    },
    {
      "auxiliary_loss_clip": 0.01182416,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.05889058,
      "balance_loss_mlp": 1.02110696,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.8999544267433428,
      "language_loss": 0.84315413,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86526805,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.612750291824341
    },
    {
      "auxiliary_loss_clip": 0.01173073,
      "auxiliary_loss_mlp": 0.01032383,
      "balance_loss_clip": 1.05980992,
      "balance_loss_mlp": 1.02366924,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 1.8791105787396443,
      "language_loss": 0.69597912,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.71803367,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.7640531063079834
    },
    {
      "auxiliary_loss_clip": 0.01168677,
      "auxiliary_loss_mlp": 0.00901925,
      "balance_loss_clip": 1.05728471,
      "balance_loss_mlp": 1.00072014,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.2366714541650436,
      "language_loss": 0.73352802,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.75423396,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.5833802223205566
    },
    {
      "auxiliary_loss_clip": 0.01196647,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.05998814,
      "balance_loss_mlp": 1.02277684,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.0651504322050585,
      "language_loss": 0.85152364,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.87380505,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 3.52064847946167
    },
    {
      "auxiliary_loss_clip": 0.01178318,
      "auxiliary_loss_mlp": 0.01030484,
      "balance_loss_clip": 1.05450594,
      "balance_loss_mlp": 1.02194905,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.782122906162293,
      "language_loss": 0.8847127,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90680069,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.6554691791534424
    },
    {
      "auxiliary_loss_clip": 0.01179145,
      "auxiliary_loss_mlp": 0.01035455,
      "balance_loss_clip": 1.05554855,
      "balance_loss_mlp": 1.026824,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.203237555553484,
      "language_loss": 0.8519783,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87412429,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 3.366312026977539
    },
    {
      "auxiliary_loss_clip": 0.01189466,
      "auxiliary_loss_mlp": 0.01039413,
      "balance_loss_clip": 1.05922151,
      "balance_loss_mlp": 1.03147388,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.1044550072607398,
      "language_loss": 0.78976417,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81205297,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.5694046020507812
    },
    {
      "auxiliary_loss_clip": 0.01175438,
      "auxiliary_loss_mlp": 0.00901873,
      "balance_loss_clip": 1.0566349,
      "balance_loss_mlp": 1.0007149,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.7984608429446147,
      "language_loss": 0.81125695,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.83203006,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.586859703063965
    },
    {
      "auxiliary_loss_clip": 0.01163872,
      "auxiliary_loss_mlp": 0.01031161,
      "balance_loss_clip": 1.05157089,
      "balance_loss_mlp": 1.02264404,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 4.008488275219906,
      "language_loss": 0.75419182,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77614218,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 3.559892416000366
    },
    {
      "auxiliary_loss_clip": 0.01193708,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.05715418,
      "balance_loss_mlp": 1.02233338,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.6006579818635815,
      "language_loss": 0.86404955,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.88628662,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.684612274169922
    },
    {
      "auxiliary_loss_clip": 0.0119436,
      "auxiliary_loss_mlp": 0.01033824,
      "balance_loss_clip": 1.0576005,
      "balance_loss_mlp": 1.02457678,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.9483009898458272,
      "language_loss": 0.85379434,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.87607616,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.6404263973236084
    },
    {
      "auxiliary_loss_clip": 0.01195702,
      "auxiliary_loss_mlp": 0.01028737,
      "balance_loss_clip": 1.05686224,
      "balance_loss_mlp": 1.01975441,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 2.1173339999658762,
      "language_loss": 0.73133415,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75357854,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.572619915008545
    },
    {
      "auxiliary_loss_clip": 0.01191959,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.05836141,
      "balance_loss_mlp": 1.02253437,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.164437638709647,
      "language_loss": 0.74514288,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76737463,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.6714797019958496
    },
    {
      "auxiliary_loss_clip": 0.01172819,
      "auxiliary_loss_mlp": 0.0103317,
      "balance_loss_clip": 1.05679512,
      "balance_loss_mlp": 1.02492046,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.254106113150277,
      "language_loss": 0.80862677,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83068669,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.6282546520233154
    },
    {
      "auxiliary_loss_clip": 0.0119572,
      "auxiliary_loss_mlp": 0.01033068,
      "balance_loss_clip": 1.06143582,
      "balance_loss_mlp": 1.0248487,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 2.1909167294495937,
      "language_loss": 0.74651104,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76879895,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.538196325302124
    },
    {
      "auxiliary_loss_clip": 0.0118289,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.05536985,
      "balance_loss_mlp": 1.01980817,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 2.046661436582203,
      "language_loss": 0.80091965,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82303095,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.733990430831909
    },
    {
      "auxiliary_loss_clip": 0.01164334,
      "auxiliary_loss_mlp": 0.01035223,
      "balance_loss_clip": 1.05405116,
      "balance_loss_mlp": 1.02635384,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 2.392145326206537,
      "language_loss": 0.76622736,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78822297,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.6988720893859863
    },
    {
      "auxiliary_loss_clip": 0.01197918,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.06053281,
      "balance_loss_mlp": 1.02040553,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 2.203094078388456,
      "language_loss": 0.74000525,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76226544,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.6176557540893555
    },
    {
      "auxiliary_loss_clip": 0.0115795,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.04891515,
      "balance_loss_mlp": 1.02046394,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.8718720917885763,
      "language_loss": 0.83594114,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.8578068,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.642662763595581
    },
    {
      "auxiliary_loss_clip": 0.01077821,
      "auxiliary_loss_mlp": 0.01002076,
      "balance_loss_clip": 1.02638435,
      "balance_loss_mlp": 1.00084198,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7886596224810256,
      "language_loss": 0.5477972,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56859612,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.192556619644165
    },
    {
      "auxiliary_loss_clip": 0.01184549,
      "auxiliary_loss_mlp": 0.01032938,
      "balance_loss_clip": 1.05877042,
      "balance_loss_mlp": 1.0248642,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 3.4622748883337517,
      "language_loss": 0.76422095,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78639591,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.660416841506958
    },
    {
      "auxiliary_loss_clip": 0.01188918,
      "auxiliary_loss_mlp": 0.01026953,
      "balance_loss_clip": 1.05887938,
      "balance_loss_mlp": 1.01810718,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.6870268417049095,
      "language_loss": 0.71345711,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73561585,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.645899534225464
    },
    {
      "auxiliary_loss_clip": 0.01170459,
      "auxiliary_loss_mlp": 0.01031844,
      "balance_loss_clip": 1.04934931,
      "balance_loss_mlp": 1.02379727,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.682044602534643,
      "language_loss": 0.82241821,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84444124,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.7374634742736816
    },
    {
      "auxiliary_loss_clip": 0.01181456,
      "auxiliary_loss_mlp": 0.01033423,
      "balance_loss_clip": 1.05335116,
      "balance_loss_mlp": 1.02522135,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.217562868542639,
      "language_loss": 0.85287637,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.87502515,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.7536818981170654
    },
    {
      "auxiliary_loss_clip": 0.01163899,
      "auxiliary_loss_mlp": 0.01036603,
      "balance_loss_clip": 1.0506686,
      "balance_loss_mlp": 1.02781749,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 2.2409598542260922,
      "language_loss": 0.73444384,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75644886,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.746495485305786
    },
    {
      "auxiliary_loss_clip": 0.01182264,
      "auxiliary_loss_mlp": 0.01027446,
      "balance_loss_clip": 1.05570245,
      "balance_loss_mlp": 1.01904178,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 1.7747329433055166,
      "language_loss": 0.85236138,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87445855,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 3.5136148929595947
    },
    {
      "auxiliary_loss_clip": 0.01156908,
      "auxiliary_loss_mlp": 0.01030053,
      "balance_loss_clip": 1.0509727,
      "balance_loss_mlp": 1.02169621,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.547782468650543,
      "language_loss": 0.80581456,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82768422,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.7874717712402344
    },
    {
      "auxiliary_loss_clip": 0.01168045,
      "auxiliary_loss_mlp": 0.01031502,
      "balance_loss_clip": 1.05078471,
      "balance_loss_mlp": 1.02259147,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.147681974350917,
      "language_loss": 0.7418623,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76385778,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.6386876106262207
    },
    {
      "auxiliary_loss_clip": 0.01196106,
      "auxiliary_loss_mlp": 0.01030932,
      "balance_loss_clip": 1.05797458,
      "balance_loss_mlp": 1.02205074,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 1.9641303495805473,
      "language_loss": 0.77834296,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80061328,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.6165688037872314
    },
    {
      "auxiliary_loss_clip": 0.01184341,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.05466628,
      "balance_loss_mlp": 1.01902652,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.9121281314640381,
      "language_loss": 0.76632869,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.78844732,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.607841730117798
    },
    {
      "auxiliary_loss_clip": 0.01096326,
      "auxiliary_loss_mlp": 0.01004309,
      "balance_loss_clip": 1.02637446,
      "balance_loss_mlp": 1.00308692,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7263518396539312,
      "language_loss": 0.56654036,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58754671,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 3.9091243743896484
    },
    {
      "auxiliary_loss_clip": 0.01176318,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.05394864,
      "balance_loss_mlp": 1.01818109,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.5232080200132083,
      "language_loss": 0.81997353,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84200501,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 3.569650888442993
    },
    {
      "auxiliary_loss_clip": 0.01169996,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.05015934,
      "balance_loss_mlp": 1.02360225,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.6717370338369097,
      "language_loss": 0.71651268,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.73852909,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.736642360687256
    },
    {
      "auxiliary_loss_clip": 0.0118744,
      "auxiliary_loss_mlp": 0.01030131,
      "balance_loss_clip": 1.05714786,
      "balance_loss_mlp": 1.02189374,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 5.1740610817429165,
      "language_loss": 0.76779276,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78996849,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.593794345855713
    },
    {
      "auxiliary_loss_clip": 0.01189193,
      "auxiliary_loss_mlp": 0.01029923,
      "balance_loss_clip": 1.06190574,
      "balance_loss_mlp": 1.02127409,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.650533346235286,
      "language_loss": 0.83793938,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.86013049,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.6017956733703613
    },
    {
      "auxiliary_loss_clip": 0.01169199,
      "auxiliary_loss_mlp": 0.00901662,
      "balance_loss_clip": 1.05315876,
      "balance_loss_mlp": 1.00076151,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.9954163718994025,
      "language_loss": 0.77840924,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.79911786,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 3.6019883155822754
    },
    {
      "auxiliary_loss_clip": 0.01180085,
      "auxiliary_loss_mlp": 0.00901211,
      "balance_loss_clip": 1.05878234,
      "balance_loss_mlp": 1.00068498,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 2.6048867512045417,
      "language_loss": 0.87004983,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.89086276,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.6457033157348633
    },
    {
      "auxiliary_loss_clip": 0.01175315,
      "auxiliary_loss_mlp": 0.01032033,
      "balance_loss_clip": 1.05367172,
      "balance_loss_mlp": 1.02297306,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.2575393315114485,
      "language_loss": 0.75732356,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.77939701,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.6022207736968994
    },
    {
      "auxiliary_loss_clip": 0.01136947,
      "auxiliary_loss_mlp": 0.01028368,
      "balance_loss_clip": 1.04922998,
      "balance_loss_mlp": 1.01977313,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.80751789568547,
      "language_loss": 0.82774019,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.84939337,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.8105359077453613
    },
    {
      "auxiliary_loss_clip": 0.01073123,
      "auxiliary_loss_mlp": 0.01001787,
      "balance_loss_clip": 1.02700138,
      "balance_loss_mlp": 1.00053561,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7679289733002298,
      "language_loss": 0.6034261,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62417519,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.3186278343200684
    },
    {
      "auxiliary_loss_clip": 0.01164346,
      "auxiliary_loss_mlp": 0.01031513,
      "balance_loss_clip": 1.05034304,
      "balance_loss_mlp": 1.02252412,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 2.210921124507439,
      "language_loss": 0.76970404,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79166269,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.668328046798706
    },
    {
      "auxiliary_loss_clip": 0.01188437,
      "auxiliary_loss_mlp": 0.01034316,
      "balance_loss_clip": 1.05692697,
      "balance_loss_mlp": 1.02563155,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.7377609866341646,
      "language_loss": 0.78162289,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80385041,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.698367118835449
    },
    {
      "auxiliary_loss_clip": 0.01155906,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.05005038,
      "balance_loss_mlp": 1.02434134,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.1891566153951523,
      "language_loss": 0.69605279,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.71793914,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.71238112449646
    },
    {
      "auxiliary_loss_clip": 0.01160819,
      "auxiliary_loss_mlp": 0.01035845,
      "balance_loss_clip": 1.04945242,
      "balance_loss_mlp": 1.02687502,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.6054675613169826,
      "language_loss": 0.78396809,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80593479,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.7891664505004883
    },
    {
      "auxiliary_loss_clip": 0.01192693,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.05581939,
      "balance_loss_mlp": 1.01682973,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 1.921428835564511,
      "language_loss": 0.83685541,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85903507,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.5696935653686523
    },
    {
      "auxiliary_loss_clip": 0.01176707,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.05322027,
      "balance_loss_mlp": 1.02166378,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 2.2553188629124032,
      "language_loss": 0.69539458,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71746701,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.6964809894561768
    },
    {
      "auxiliary_loss_clip": 0.01181264,
      "auxiliary_loss_mlp": 0.01030713,
      "balance_loss_clip": 1.05741096,
      "balance_loss_mlp": 1.02260065,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.4029672883517543,
      "language_loss": 0.70286089,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72498071,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.6690874099731445
    },
    {
      "auxiliary_loss_clip": 0.01177261,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.0541563,
      "balance_loss_mlp": 1.02535701,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 2.0002805375459274,
      "language_loss": 0.77283156,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79494762,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.652013063430786
    },
    {
      "auxiliary_loss_clip": 0.01075651,
      "auxiliary_loss_mlp": 0.0100158,
      "balance_loss_clip": 1.02239835,
      "balance_loss_mlp": 1.0004952,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8374621905578953,
      "language_loss": 0.65129077,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67206311,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.2990260124206543
    },
    {
      "auxiliary_loss_clip": 0.01186002,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.05461311,
      "balance_loss_mlp": 1.01784492,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.7740081423520668,
      "language_loss": 0.75500518,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.77712023,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.5858497619628906
    },
    {
      "auxiliary_loss_clip": 0.01150845,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.04996204,
      "balance_loss_mlp": 1.01789355,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.236679427729955,
      "language_loss": 0.89454818,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91631812,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.6720168590545654
    },
    {
      "auxiliary_loss_clip": 0.0115481,
      "auxiliary_loss_mlp": 0.01024039,
      "balance_loss_clip": 1.04961503,
      "balance_loss_mlp": 1.01628399,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 2.0266144897953007,
      "language_loss": 0.81310791,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83489639,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.7010343074798584
    },
    {
      "auxiliary_loss_clip": 0.01193796,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.06016958,
      "balance_loss_mlp": 1.02438354,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.86206656133781,
      "language_loss": 0.69275367,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71501899,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 3.5949344635009766
    },
    {
      "auxiliary_loss_clip": 0.0113756,
      "auxiliary_loss_mlp": 0.01027222,
      "balance_loss_clip": 1.05333138,
      "balance_loss_mlp": 1.0198462,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 1.97558046809107,
      "language_loss": 0.79636675,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.8180145,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.7313156127929688
    },
    {
      "auxiliary_loss_clip": 0.01159541,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.05332899,
      "balance_loss_mlp": 1.01605749,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.089225019387508,
      "language_loss": 0.69019198,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71203059,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.750480890274048
    },
    {
      "auxiliary_loss_clip": 0.01173717,
      "auxiliary_loss_mlp": 0.01035328,
      "balance_loss_clip": 1.05619431,
      "balance_loss_mlp": 1.0275197,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 4.133295072642883,
      "language_loss": 0.70059705,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.7226876,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.689209461212158
    },
    {
      "auxiliary_loss_clip": 0.0118729,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.0560801,
      "balance_loss_mlp": 1.01781487,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.757860340437579,
      "language_loss": 0.8400985,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.86222982,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 3.6046855449676514
    },
    {
      "auxiliary_loss_clip": 0.01190759,
      "auxiliary_loss_mlp": 0.01023567,
      "balance_loss_clip": 1.05954909,
      "balance_loss_mlp": 1.01641393,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 1.6660506187844093,
      "language_loss": 0.87998414,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.90212739,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 3.524218797683716
    },
    {
      "auxiliary_loss_clip": 0.01159032,
      "auxiliary_loss_mlp": 0.01033312,
      "balance_loss_clip": 1.05188203,
      "balance_loss_mlp": 1.02450824,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 2.243041315279228,
      "language_loss": 0.9164989,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93842232,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.6544477939605713
    },
    {
      "auxiliary_loss_clip": 0.01164597,
      "auxiliary_loss_mlp": 0.01032579,
      "balance_loss_clip": 1.05106151,
      "balance_loss_mlp": 1.02371001,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 3.12832822754562,
      "language_loss": 0.67399812,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.69596994,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.637582778930664
    },
    {
      "auxiliary_loss_clip": 0.01192517,
      "auxiliary_loss_mlp": 0.01027745,
      "balance_loss_clip": 1.05826712,
      "balance_loss_mlp": 1.01975214,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.5899571776822792,
      "language_loss": 0.78011656,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80231917,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.6060550212860107
    },
    {
      "auxiliary_loss_clip": 0.01190877,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.05712628,
      "balance_loss_mlp": 1.01994705,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 2.570952771468358,
      "language_loss": 0.7030195,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.72521091,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 3.498246192932129
    },
    {
      "auxiliary_loss_clip": 0.01177513,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.05387092,
      "balance_loss_mlp": 1.02075374,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.175071959892652,
      "language_loss": 0.67586541,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69793355,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.7026185989379883
    },
    {
      "auxiliary_loss_clip": 0.01173683,
      "auxiliary_loss_mlp": 0.00901096,
      "balance_loss_clip": 1.05455256,
      "balance_loss_mlp": 1.00088978,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 2.0206582347342623,
      "language_loss": 0.65123671,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.67198443,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.7225918769836426
    },
    {
      "auxiliary_loss_clip": 0.01173759,
      "auxiliary_loss_mlp": 0.01038413,
      "balance_loss_clip": 1.05231285,
      "balance_loss_mlp": 1.0295198,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 2.458547364148317,
      "language_loss": 0.87281764,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.89493936,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.720667839050293
    },
    {
      "auxiliary_loss_clip": 0.01167842,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.05204344,
      "balance_loss_mlp": 1.02097046,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 1.902919620353696,
      "language_loss": 0.7856909,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80766416,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.654114007949829
    },
    {
      "auxiliary_loss_clip": 0.01173827,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.05249143,
      "balance_loss_mlp": 1.01858997,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.0940101945396803,
      "language_loss": 0.8502779,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87228942,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.677415132522583
    },
    {
      "auxiliary_loss_clip": 0.01167407,
      "auxiliary_loss_mlp": 0.01023453,
      "balance_loss_clip": 1.05620408,
      "balance_loss_mlp": 1.01620471,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 1.8999734425832469,
      "language_loss": 0.85444731,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87635589,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.698500394821167
    },
    {
      "auxiliary_loss_clip": 0.01167276,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.05448139,
      "balance_loss_mlp": 1.01407981,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 3.048175566975522,
      "language_loss": 0.82225859,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84415138,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.755310535430908
    },
    {
      "auxiliary_loss_clip": 0.01184602,
      "auxiliary_loss_mlp": 0.01026398,
      "balance_loss_clip": 1.05639338,
      "balance_loss_mlp": 1.01802945,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 1.8118696356967807,
      "language_loss": 0.79892159,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82103157,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.7474303245544434
    },
    {
      "auxiliary_loss_clip": 0.01185951,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.05913162,
      "balance_loss_mlp": 1.02201438,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 7.239039866397316,
      "language_loss": 0.79379547,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81595832,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.6111912727355957
    },
    {
      "auxiliary_loss_clip": 0.01163936,
      "auxiliary_loss_mlp": 0.01033117,
      "balance_loss_clip": 1.05590916,
      "balance_loss_mlp": 1.02414632,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 2.1524736890680836,
      "language_loss": 0.68717062,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.70914114,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.692230463027954
    },
    {
      "auxiliary_loss_clip": 0.01191922,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.06231284,
      "balance_loss_mlp": 1.02072191,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.5226564244571228,
      "language_loss": 0.77342784,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79563224,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.650993824005127
    },
    {
      "auxiliary_loss_clip": 0.01153984,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.05048096,
      "balance_loss_mlp": 1.02251101,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.6743060237796614,
      "language_loss": 0.59492552,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.61677158,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.640664577484131
    },
    {
      "auxiliary_loss_clip": 0.01193481,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.06020284,
      "balance_loss_mlp": 1.02428913,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 2.534196701107542,
      "language_loss": 0.73518282,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75744003,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.6254966259002686
    },
    {
      "auxiliary_loss_clip": 0.01156787,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.05348253,
      "balance_loss_mlp": 1.01926208,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.188783921965356,
      "language_loss": 0.75364655,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77549034,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.683185338973999
    },
    {
      "auxiliary_loss_clip": 0.01185781,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.05621886,
      "balance_loss_mlp": 1.01678467,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 1.8911496865359145,
      "language_loss": 0.85447896,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87657738,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.619330883026123
    },
    {
      "auxiliary_loss_clip": 0.01169129,
      "auxiliary_loss_mlp": 0.01026757,
      "balance_loss_clip": 1.05296123,
      "balance_loss_mlp": 1.0192883,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.778800539348267,
      "language_loss": 0.74965429,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.77161312,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.7900707721710205
    },
    {
      "auxiliary_loss_clip": 0.01165728,
      "auxiliary_loss_mlp": 0.01021722,
      "balance_loss_clip": 1.05758929,
      "balance_loss_mlp": 1.0134728,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 3.1614310351506774,
      "language_loss": 0.81819916,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84007365,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 3.6940672397613525
    },
    {
      "auxiliary_loss_clip": 0.01175609,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.05741394,
      "balance_loss_mlp": 1.02355671,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.2573277844983326,
      "language_loss": 0.75354195,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77561909,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.718527317047119
    },
    {
      "auxiliary_loss_clip": 0.01195242,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.06034446,
      "balance_loss_mlp": 1.01972973,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.582673498865637,
      "language_loss": 0.83025384,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85248411,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.6312825679779053
    },
    {
      "auxiliary_loss_clip": 0.01173496,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.05481005,
      "balance_loss_mlp": 1.01914096,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.5078904698878386,
      "language_loss": 0.84143168,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86343491,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.6717636585235596
    },
    {
      "auxiliary_loss_clip": 0.01151809,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.05279958,
      "balance_loss_mlp": 1.01857948,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 2.236993675755018,
      "language_loss": 0.72772574,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.74951422,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.7352631092071533
    },
    {
      "auxiliary_loss_clip": 0.0119537,
      "auxiliary_loss_mlp": 0.01031088,
      "balance_loss_clip": 1.06002402,
      "balance_loss_mlp": 1.02316058,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.0303212714037575,
      "language_loss": 0.75035626,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77262092,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 3.528353452682495
    },
    {
      "auxiliary_loss_clip": 0.01194501,
      "auxiliary_loss_mlp": 0.01032272,
      "balance_loss_clip": 1.05843008,
      "balance_loss_mlp": 1.02370083,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 1.7130954129863858,
      "language_loss": 0.71707058,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73933828,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 3.617539405822754
    },
    {
      "auxiliary_loss_clip": 0.01189861,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 1.0589304,
      "balance_loss_mlp": 1.02006221,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 3.2217534449087193,
      "language_loss": 0.75968534,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78186715,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.698864698410034
    },
    {
      "auxiliary_loss_clip": 0.01156972,
      "auxiliary_loss_mlp": 0.01028241,
      "balance_loss_clip": 1.0551616,
      "balance_loss_mlp": 1.02057576,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.7393967195579798,
      "language_loss": 0.81723356,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.8390857,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.7481870651245117
    },
    {
      "auxiliary_loss_clip": 0.01193243,
      "auxiliary_loss_mlp": 0.01025571,
      "balance_loss_clip": 1.05890799,
      "balance_loss_mlp": 1.01766753,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 2.718222628967824,
      "language_loss": 0.75231665,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77450478,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.582554340362549
    },
    {
      "auxiliary_loss_clip": 0.01187586,
      "auxiliary_loss_mlp": 0.01030867,
      "balance_loss_clip": 1.0614562,
      "balance_loss_mlp": 1.02285635,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.803393901375735,
      "language_loss": 0.80373633,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82592088,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 3.5713560581207275
    },
    {
      "auxiliary_loss_clip": 0.011757,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.05607033,
      "balance_loss_mlp": 1.01758492,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.7107995581002917,
      "language_loss": 0.73444986,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75645339,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.6750247478485107
    },
    {
      "auxiliary_loss_clip": 0.01192179,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.05815554,
      "balance_loss_mlp": 1.02184224,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.041100937444051,
      "language_loss": 0.71066326,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73288,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.680569648742676
    },
    {
      "auxiliary_loss_clip": 0.01180416,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 1.05349362,
      "balance_loss_mlp": 1.02127039,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 3.1966422512601014,
      "language_loss": 0.75670707,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77880251,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.5880143642425537
    },
    {
      "auxiliary_loss_clip": 0.01177952,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.05887437,
      "balance_loss_mlp": 1.01796222,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.7965310415645257,
      "language_loss": 0.86161345,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.8836503,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.7898499965667725
    },
    {
      "auxiliary_loss_clip": 0.01170206,
      "auxiliary_loss_mlp": 0.01030314,
      "balance_loss_clip": 1.05613863,
      "balance_loss_mlp": 1.02162373,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 3.184209814770639,
      "language_loss": 0.71828747,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.74029267,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.7072277069091797
    },
    {
      "auxiliary_loss_clip": 0.01177933,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05702913,
      "balance_loss_mlp": 1.01909637,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.103995230320372,
      "language_loss": 0.77172393,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79377431,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.6915979385375977
    },
    {
      "auxiliary_loss_clip": 0.01158765,
      "auxiliary_loss_mlp": 0.01027336,
      "balance_loss_clip": 1.0486176,
      "balance_loss_mlp": 1.01985514,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.123359176465508,
      "language_loss": 0.83098745,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.85284841,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.6407108306884766
    },
    {
      "auxiliary_loss_clip": 0.01192211,
      "auxiliary_loss_mlp": 0.01035511,
      "balance_loss_clip": 1.06114912,
      "balance_loss_mlp": 1.02778029,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 2.1756355600945643,
      "language_loss": 0.65281874,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67509592,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.675856351852417
    },
    {
      "auxiliary_loss_clip": 0.01183356,
      "auxiliary_loss_mlp": 0.01030622,
      "balance_loss_clip": 1.05798292,
      "balance_loss_mlp": 1.02282882,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.8177797724764315,
      "language_loss": 0.6808812,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.70302093,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.584333896636963
    },
    {
      "auxiliary_loss_clip": 0.0114856,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.05001235,
      "balance_loss_mlp": 1.01843905,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.505845093393236,
      "language_loss": 0.76407933,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.78583193,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.7556464672088623
    },
    {
      "auxiliary_loss_clip": 0.0115043,
      "auxiliary_loss_mlp": 0.00901878,
      "balance_loss_clip": 1.04926801,
      "balance_loss_mlp": 1.00120211,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.690963980671044,
      "language_loss": 0.75059485,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.77111799,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.6457533836364746
    },
    {
      "auxiliary_loss_clip": 0.01152884,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.04947221,
      "balance_loss_mlp": 1.02231979,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.9098494130144,
      "language_loss": 0.74608147,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76791549,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.6902010440826416
    },
    {
      "auxiliary_loss_clip": 0.01168668,
      "auxiliary_loss_mlp": 0.01023036,
      "balance_loss_clip": 1.05253482,
      "balance_loss_mlp": 1.01510835,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 2.535412551257474,
      "language_loss": 0.87242401,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89434105,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.6609723567962646
    },
    {
      "auxiliary_loss_clip": 0.01127656,
      "auxiliary_loss_mlp": 0.01031648,
      "balance_loss_clip": 1.0460012,
      "balance_loss_mlp": 1.02355409,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 2.203387709356694,
      "language_loss": 0.79904264,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.82063568,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.812556028366089
    },
    {
      "auxiliary_loss_clip": 0.0118417,
      "auxiliary_loss_mlp": 0.01029385,
      "balance_loss_clip": 1.05676031,
      "balance_loss_mlp": 1.02113557,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 2.6266694863665734,
      "language_loss": 0.7846815,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80681705,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.6239094734191895
    },
    {
      "auxiliary_loss_clip": 0.01187209,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.05797589,
      "balance_loss_mlp": 1.02076924,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 1.980777053778119,
      "language_loss": 0.7047224,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.72688174,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.6985714435577393
    },
    {
      "auxiliary_loss_clip": 0.01174271,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.05513501,
      "balance_loss_mlp": 1.0228231,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 2.4519561036652697,
      "language_loss": 0.82959867,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85164821,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 3.6054539680480957
    },
    {
      "auxiliary_loss_clip": 0.0117145,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.05498195,
      "balance_loss_mlp": 1.02258563,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.6857789107059244,
      "language_loss": 0.75610238,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77812445,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.61464786529541
    },
    {
      "auxiliary_loss_clip": 0.01193827,
      "auxiliary_loss_mlp": 0.00901154,
      "balance_loss_clip": 1.06015825,
      "balance_loss_mlp": 1.00116181,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 2.0433390843010515,
      "language_loss": 0.66776496,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.68871474,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.6377904415130615
    },
    {
      "auxiliary_loss_clip": 0.01182169,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.05600989,
      "balance_loss_mlp": 1.02340102,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 4.8476869161316305,
      "language_loss": 0.68481356,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70695198,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.7222208976745605
    },
    {
      "auxiliary_loss_clip": 0.01191965,
      "auxiliary_loss_mlp": 0.01028801,
      "balance_loss_clip": 1.05817759,
      "balance_loss_mlp": 1.02047443,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 5.3554985984525825,
      "language_loss": 0.61298323,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63519084,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 3.60900616645813
    },
    {
      "auxiliary_loss_clip": 0.01138361,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.04835296,
      "balance_loss_mlp": 1.02008486,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 2.0701151729609397,
      "language_loss": 0.79283237,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81449163,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.7982451915740967
    },
    {
      "auxiliary_loss_clip": 0.01180436,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.05749917,
      "balance_loss_mlp": 1.02102721,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.8011902556555983,
      "language_loss": 0.75076175,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77285719,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 3.585975170135498
    },
    {
      "auxiliary_loss_clip": 0.01187533,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.05715227,
      "balance_loss_mlp": 1.02230537,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.4521215207825042,
      "language_loss": 0.68559939,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70778215,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 2.9920144081115723
    },
    {
      "auxiliary_loss_clip": 0.01152395,
      "auxiliary_loss_mlp": 0.01027128,
      "balance_loss_clip": 1.05022895,
      "balance_loss_mlp": 1.01953733,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.8141431573759674,
      "language_loss": 0.88421869,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90601385,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.645688772201538
    },
    {
      "auxiliary_loss_clip": 0.0118443,
      "auxiliary_loss_mlp": 0.01030725,
      "balance_loss_clip": 1.05785942,
      "balance_loss_mlp": 1.02225566,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.2050201083538625,
      "language_loss": 0.71550399,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73765552,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.584500312805176
    },
    {
      "auxiliary_loss_clip": 0.01148059,
      "auxiliary_loss_mlp": 0.00902594,
      "balance_loss_clip": 1.04701567,
      "balance_loss_mlp": 1.00129735,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 3.4318951489851353,
      "language_loss": 0.83577383,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.85628033,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 3.6821794509887695
    },
    {
      "auxiliary_loss_clip": 0.01194777,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 1.06008053,
      "balance_loss_mlp": 1.02116728,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 4.071474701560841,
      "language_loss": 0.91913533,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94137454,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.5261242389678955
    },
    {
      "auxiliary_loss_clip": 0.01168903,
      "auxiliary_loss_mlp": 0.01021369,
      "balance_loss_clip": 1.05422139,
      "balance_loss_mlp": 1.01372194,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 2.0862662274427555,
      "language_loss": 0.74547601,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76737881,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.72602915763855
    },
    {
      "auxiliary_loss_clip": 0.01182769,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.0573281,
      "balance_loss_mlp": 1.01763558,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.495253190167636,
      "language_loss": 0.71602345,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73810697,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.5699145793914795
    },
    {
      "auxiliary_loss_clip": 0.01187961,
      "auxiliary_loss_mlp": 0.01022498,
      "balance_loss_clip": 1.0562253,
      "balance_loss_mlp": 1.01450562,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 3.7241070340003155,
      "language_loss": 0.76741743,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78952205,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.666367769241333
    },
    {
      "auxiliary_loss_clip": 0.01195055,
      "auxiliary_loss_mlp": 0.01032805,
      "balance_loss_clip": 1.05865312,
      "balance_loss_mlp": 1.02406144,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.5748482241588757,
      "language_loss": 0.80949306,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.83177161,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.530468702316284
    },
    {
      "auxiliary_loss_clip": 0.01183394,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 1.06006837,
      "balance_loss_mlp": 1.01822293,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.7506211109638183,
      "language_loss": 0.82690358,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.8489964,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.648489236831665
    },
    {
      "auxiliary_loss_clip": 0.01160972,
      "auxiliary_loss_mlp": 0.0102479,
      "balance_loss_clip": 1.05186701,
      "balance_loss_mlp": 1.01728034,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.2303850331041244,
      "language_loss": 0.7633822,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78523982,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.6172049045562744
    },
    {
      "auxiliary_loss_clip": 0.0119272,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.06016731,
      "balance_loss_mlp": 1.01839864,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.582622458473811,
      "language_loss": 0.72952843,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75171071,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.7242071628570557
    },
    {
      "auxiliary_loss_clip": 0.01195312,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.05849552,
      "balance_loss_mlp": 1.01883531,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.3245998728681565,
      "language_loss": 0.81731403,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.83954072,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.5788378715515137
    },
    {
      "auxiliary_loss_clip": 0.0119665,
      "auxiliary_loss_mlp": 0.01037366,
      "balance_loss_clip": 1.05990303,
      "balance_loss_mlp": 1.02927208,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.9804149582558168,
      "language_loss": 0.84623742,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86857754,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.650946617126465
    },
    {
      "auxiliary_loss_clip": 0.0117427,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 1.05512607,
      "balance_loss_mlp": 1.01984513,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.8563445081393668,
      "language_loss": 0.82867277,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.85068834,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.652587413787842
    },
    {
      "auxiliary_loss_clip": 0.01170412,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.0528028,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 2.1033183113871945,
      "language_loss": 0.73051798,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75248641,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.7172231674194336
    },
    {
      "auxiliary_loss_clip": 0.01174847,
      "auxiliary_loss_mlp": 0.01027671,
      "balance_loss_clip": 1.05789685,
      "balance_loss_mlp": 1.01985693,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.1122256013544147,
      "language_loss": 0.82641482,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84843999,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.6361782550811768
    },
    {
      "auxiliary_loss_clip": 0.01186377,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.05969191,
      "balance_loss_mlp": 1.0234549,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.693445827888736,
      "language_loss": 0.75887054,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.78104484,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.604285478591919
    },
    {
      "auxiliary_loss_clip": 0.01138814,
      "auxiliary_loss_mlp": 0.01035538,
      "balance_loss_clip": 1.04589391,
      "balance_loss_mlp": 1.02660954,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 1.70308791762144,
      "language_loss": 0.76259136,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.7843349,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.8211307525634766
    },
    {
      "auxiliary_loss_clip": 0.01175925,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.05587268,
      "balance_loss_mlp": 1.02451801,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.783984761700255,
      "language_loss": 0.69031864,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.7123996,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 3.578131675720215
    },
    {
      "auxiliary_loss_clip": 0.01194537,
      "auxiliary_loss_mlp": 0.01034146,
      "balance_loss_clip": 1.06044245,
      "balance_loss_mlp": 1.02647519,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 1.8180722691637508,
      "language_loss": 0.75832427,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78061104,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.744174003601074
    },
    {
      "auxiliary_loss_clip": 0.01173641,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.05294085,
      "balance_loss_mlp": 1.01870489,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.800268725173649,
      "language_loss": 0.72856337,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.7505697,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.6813387870788574
    },
    {
      "auxiliary_loss_clip": 0.01185574,
      "auxiliary_loss_mlp": 0.0102575,
      "balance_loss_clip": 1.05478406,
      "balance_loss_mlp": 1.01810908,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.3373124348389926,
      "language_loss": 0.75003701,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.77215028,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.6368062496185303
    },
    {
      "auxiliary_loss_clip": 0.01186315,
      "auxiliary_loss_mlp": 0.01023922,
      "balance_loss_clip": 1.0583477,
      "balance_loss_mlp": 1.01629829,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.3212655248183025,
      "language_loss": 0.72987509,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.7519775,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.616652488708496
    },
    {
      "auxiliary_loss_clip": 0.01084292,
      "auxiliary_loss_mlp": 0.01012207,
      "balance_loss_clip": 1.028157,
      "balance_loss_mlp": 1.01055038,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.726719524212097,
      "language_loss": 0.61619133,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63715631,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 4.206485271453857
    },
    {
      "auxiliary_loss_clip": 0.01192608,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.05769706,
      "balance_loss_mlp": 1.01836371,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.9693875767928244,
      "language_loss": 0.86564088,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88782287,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.609114408493042
    },
    {
      "auxiliary_loss_clip": 0.01160413,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.05215359,
      "balance_loss_mlp": 1.0217638,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 2.1012240595694407,
      "language_loss": 0.87226462,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89416373,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 3.708954334259033
    },
    {
      "auxiliary_loss_clip": 0.01166058,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.05353117,
      "balance_loss_mlp": 1.02119017,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 2.278019963129649,
      "language_loss": 0.81787199,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.83982766,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.679497480392456
    },
    {
      "auxiliary_loss_clip": 0.01159736,
      "auxiliary_loss_mlp": 0.01030403,
      "balance_loss_clip": 1.054492,
      "balance_loss_mlp": 1.02304769,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.8604808694446995,
      "language_loss": 0.79855132,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.82045269,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.767641067504883
    },
    {
      "auxiliary_loss_clip": 0.01174326,
      "auxiliary_loss_mlp": 0.01030333,
      "balance_loss_clip": 1.05309486,
      "balance_loss_mlp": 1.02258468,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 4.6989683194008505,
      "language_loss": 0.76117063,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.78321719,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.7078042030334473
    },
    {
      "auxiliary_loss_clip": 0.01158507,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 1.05437565,
      "balance_loss_mlp": 1.01772904,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.3096552751146406,
      "language_loss": 0.75735319,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77919245,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 3.73245906829834
    },
    {
      "auxiliary_loss_clip": 0.01071154,
      "auxiliary_loss_mlp": 0.01004376,
      "balance_loss_clip": 1.02591753,
      "balance_loss_mlp": 1.00293398,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7258296477051553,
      "language_loss": 0.57558954,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59634483,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.3406975269317627
    },
    {
      "auxiliary_loss_clip": 0.01165019,
      "auxiliary_loss_mlp": 0.0103387,
      "balance_loss_clip": 1.05378318,
      "balance_loss_mlp": 1.02565646,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.3668900753178086,
      "language_loss": 0.88631809,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90830702,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.643752336502075
    },
    {
      "auxiliary_loss_clip": 0.01181896,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.05608857,
      "balance_loss_mlp": 1.01875567,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 14.082861996319807,
      "language_loss": 0.80142307,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.8235054,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.646723508834839
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.05169868,
      "balance_loss_mlp": 1.02275229,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 1.85297274930522,
      "language_loss": 0.81278974,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83471048,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.6726138591766357
    },
    {
      "auxiliary_loss_clip": 0.0116483,
      "auxiliary_loss_mlp": 0.01033029,
      "balance_loss_clip": 1.05312908,
      "balance_loss_mlp": 1.02526844,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.6355638132507146,
      "language_loss": 0.75404274,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77602136,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.712761878967285
    },
    {
      "auxiliary_loss_clip": 0.01169798,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.05230951,
      "balance_loss_mlp": 1.01479304,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.729470243505479,
      "language_loss": 0.82066119,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84257936,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.6414639949798584
    },
    {
      "auxiliary_loss_clip": 0.01179033,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.05148411,
      "balance_loss_mlp": 1.02117538,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 2.2431179784523088,
      "language_loss": 0.62643993,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.6485244,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.689286470413208
    },
    {
      "auxiliary_loss_clip": 0.01179468,
      "auxiliary_loss_mlp": 0.0102451,
      "balance_loss_clip": 1.0572145,
      "balance_loss_mlp": 1.01701772,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.0620848092148276,
      "language_loss": 0.83606231,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.85810214,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.627208948135376
    },
    {
      "auxiliary_loss_clip": 0.01181639,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.05842352,
      "balance_loss_mlp": 1.02222848,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.075022794657889,
      "language_loss": 0.72135651,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74347371,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.6264452934265137
    },
    {
      "auxiliary_loss_clip": 0.01170655,
      "auxiliary_loss_mlp": 0.01026677,
      "balance_loss_clip": 1.0552268,
      "balance_loss_mlp": 1.01990855,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.8607879603561923,
      "language_loss": 0.79495728,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81693065,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.688734769821167
    },
    {
      "auxiliary_loss_clip": 0.01189788,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.05627525,
      "balance_loss_mlp": 1.02132177,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 4.037783248795742,
      "language_loss": 0.75823796,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.78042471,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.5424604415893555
    },
    {
      "auxiliary_loss_clip": 0.01178436,
      "auxiliary_loss_mlp": 0.01025546,
      "balance_loss_clip": 1.05551982,
      "balance_loss_mlp": 1.01839924,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.8774140700269053,
      "language_loss": 0.68354058,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70558041,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.689145803451538
    },
    {
      "auxiliary_loss_clip": 0.0117067,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.05428708,
      "balance_loss_mlp": 1.02365017,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 3.302615834956319,
      "language_loss": 0.69335264,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71537578,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.637251138687134
    },
    {
      "auxiliary_loss_clip": 0.01185444,
      "auxiliary_loss_mlp": 0.0102834,
      "balance_loss_clip": 1.05749512,
      "balance_loss_mlp": 1.02016211,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.8076242987375295,
      "language_loss": 0.81279039,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83492827,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.6412436962127686
    },
    {
      "auxiliary_loss_clip": 0.01181644,
      "auxiliary_loss_mlp": 0.01033371,
      "balance_loss_clip": 1.05737484,
      "balance_loss_mlp": 1.0258491,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.411932575648702,
      "language_loss": 0.75892043,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78107059,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.6253786087036133
    },
    {
      "auxiliary_loss_clip": 0.01188281,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.05576253,
      "balance_loss_mlp": 1.02061319,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 1.957590733294937,
      "language_loss": 0.80370104,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82586199,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 3.5191712379455566
    },
    {
      "auxiliary_loss_clip": 0.01178955,
      "auxiliary_loss_mlp": 0.0103155,
      "balance_loss_clip": 1.05570817,
      "balance_loss_mlp": 1.02433777,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 2.267535475527109,
      "language_loss": 0.67524064,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69734567,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.683659791946411
    },
    {
      "auxiliary_loss_clip": 0.01180937,
      "auxiliary_loss_mlp": 0.01023493,
      "balance_loss_clip": 1.05726933,
      "balance_loss_mlp": 1.01636469,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.8718066508431006,
      "language_loss": 0.8093133,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.8313576,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.6201400756835938
    },
    {
      "auxiliary_loss_clip": 0.01168715,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.05216908,
      "balance_loss_mlp": 1.0220089,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.3287348263630827,
      "language_loss": 0.73314828,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75513077,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.6184420585632324
    },
    {
      "auxiliary_loss_clip": 0.01192719,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.0548588,
      "balance_loss_mlp": 1.01822901,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 2.9080604944458637,
      "language_loss": 0.66990495,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69209635,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 3.51488995552063
    },
    {
      "auxiliary_loss_clip": 0.01160493,
      "auxiliary_loss_mlp": 0.00901511,
      "balance_loss_clip": 1.05308533,
      "balance_loss_mlp": 1.00142813,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.2632933624900855,
      "language_loss": 0.85907185,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87969184,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.8414013385772705
    },
    {
      "auxiliary_loss_clip": 0.01192058,
      "auxiliary_loss_mlp": 0.01031341,
      "balance_loss_clip": 1.0587945,
      "balance_loss_mlp": 1.02374125,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 2.7117181589414523,
      "language_loss": 0.9050076,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92724156,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 3.488103151321411
    },
    {
      "auxiliary_loss_clip": 0.01166669,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.05450702,
      "balance_loss_mlp": 1.0235858,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.9018380616897863,
      "language_loss": 0.78666669,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.80864322,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.680304765701294
    },
    {
      "auxiliary_loss_clip": 0.01172721,
      "auxiliary_loss_mlp": 0.00901098,
      "balance_loss_clip": 1.05373704,
      "balance_loss_mlp": 1.00139713,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.2286815321601208,
      "language_loss": 0.9106673,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.93140548,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.6069416999816895
    },
    {
      "auxiliary_loss_clip": 0.01169137,
      "auxiliary_loss_mlp": 0.0102663,
      "balance_loss_clip": 1.05374599,
      "balance_loss_mlp": 1.01800513,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 1.951522679179442,
      "language_loss": 0.65935606,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68131375,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.648578643798828
    },
    {
      "auxiliary_loss_clip": 0.01171924,
      "auxiliary_loss_mlp": 0.01028436,
      "balance_loss_clip": 1.05319643,
      "balance_loss_mlp": 1.02134347,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 2.088830785769927,
      "language_loss": 0.66084588,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.68284947,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 3.561152696609497
    },
    {
      "auxiliary_loss_clip": 0.01160232,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05083799,
      "balance_loss_mlp": 1.01893091,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.1590638236521777,
      "language_loss": 0.62251669,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.64438432,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.647516965866089
    },
    {
      "auxiliary_loss_clip": 0.011855,
      "auxiliary_loss_mlp": 0.01029198,
      "balance_loss_clip": 1.05861044,
      "balance_loss_mlp": 1.0219115,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 2.8258709990673774,
      "language_loss": 0.67869496,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.70084202,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.6464107036590576
    },
    {
      "auxiliary_loss_clip": 0.01163638,
      "auxiliary_loss_mlp": 0.01030741,
      "balance_loss_clip": 1.05279124,
      "balance_loss_mlp": 1.02320647,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 2.0354870523908937,
      "language_loss": 0.7928853,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81482911,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.721165418624878
    },
    {
      "auxiliary_loss_clip": 0.01182914,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.05348682,
      "balance_loss_mlp": 1.01753819,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 1.9887749165293922,
      "language_loss": 0.6818648,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.7039541,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.6869406700134277
    },
    {
      "auxiliary_loss_clip": 0.01171265,
      "auxiliary_loss_mlp": 0.01029262,
      "balance_loss_clip": 1.05664921,
      "balance_loss_mlp": 1.02120912,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 2.1847472056996713,
      "language_loss": 0.7761842,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79818952,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.657980442047119
    },
    {
      "auxiliary_loss_clip": 0.01164038,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.05115342,
      "balance_loss_mlp": 1.01733911,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 1.9686333357644759,
      "language_loss": 0.74259192,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76448476,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.6419503688812256
    },
    {
      "auxiliary_loss_clip": 0.01159829,
      "auxiliary_loss_mlp": 0.01031759,
      "balance_loss_clip": 1.05107999,
      "balance_loss_mlp": 1.02392054,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.5786062518890411,
      "language_loss": 0.79328507,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81520098,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.794386625289917
    },
    {
      "auxiliary_loss_clip": 0.01161995,
      "auxiliary_loss_mlp": 0.0102741,
      "balance_loss_clip": 1.05108333,
      "balance_loss_mlp": 1.01919627,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 2.0754363783181744,
      "language_loss": 0.78623581,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80812985,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.7209129333496094
    },
    {
      "auxiliary_loss_clip": 0.01182411,
      "auxiliary_loss_mlp": 0.01029203,
      "balance_loss_clip": 1.05891824,
      "balance_loss_mlp": 1.0218538,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.8493488700587695,
      "language_loss": 0.81597412,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.8380903,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.6573824882507324
    },
    {
      "auxiliary_loss_clip": 0.01073348,
      "auxiliary_loss_mlp": 0.01005676,
      "balance_loss_clip": 1.02246475,
      "balance_loss_mlp": 1.00437057,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7489216996504012,
      "language_loss": 0.6045146,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62530482,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.2790040969848633
    },
    {
      "auxiliary_loss_clip": 0.01178878,
      "auxiliary_loss_mlp": 0.00901422,
      "balance_loss_clip": 1.05797052,
      "balance_loss_mlp": 1.00152957,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.0667273358591793,
      "language_loss": 0.77882892,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.79963195,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.6698412895202637
    },
    {
      "auxiliary_loss_clip": 0.01125979,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.0421735,
      "balance_loss_mlp": 1.01805687,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.7655261344732127,
      "language_loss": 0.7722227,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79374212,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.8459219932556152
    },
    {
      "auxiliary_loss_clip": 0.01181043,
      "auxiliary_loss_mlp": 0.01030109,
      "balance_loss_clip": 1.05874896,
      "balance_loss_mlp": 1.02260745,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.539347195000176,
      "language_loss": 0.76787716,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.7899887,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.6458685398101807
    },
    {
      "auxiliary_loss_clip": 0.01186949,
      "auxiliary_loss_mlp": 0.0090177,
      "balance_loss_clip": 1.05782771,
      "balance_loss_mlp": 1.00159955,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.346546010143322,
      "language_loss": 0.70005369,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.72094095,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.5539560317993164
    },
    {
      "auxiliary_loss_clip": 0.01188674,
      "auxiliary_loss_mlp": 0.01023911,
      "balance_loss_clip": 1.05783391,
      "balance_loss_mlp": 1.01632905,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 1.845949409962709,
      "language_loss": 0.84202957,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86415547,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.678593397140503
    },
    {
      "auxiliary_loss_clip": 0.01175058,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.05828345,
      "balance_loss_mlp": 1.01765239,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.2809946471497424,
      "language_loss": 0.79538232,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81738567,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.630981206893921
    },
    {
      "auxiliary_loss_clip": 0.01171326,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.05344296,
      "balance_loss_mlp": 1.01924706,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 2.37420870990666,
      "language_loss": 0.79609865,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81808209,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 3.6301968097686768
    },
    {
      "auxiliary_loss_clip": 0.01171019,
      "auxiliary_loss_mlp": 0.01025655,
      "balance_loss_clip": 1.05317175,
      "balance_loss_mlp": 1.01835907,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 2.086930216032207,
      "language_loss": 0.76024091,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78220761,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.7592427730560303
    },
    {
      "auxiliary_loss_clip": 0.01173958,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.05434418,
      "balance_loss_mlp": 1.02242875,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 2.15616734529201,
      "language_loss": 0.82224476,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84428632,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.647505521774292
    },
    {
      "auxiliary_loss_clip": 0.01166061,
      "auxiliary_loss_mlp": 0.01033535,
      "balance_loss_clip": 1.05656993,
      "balance_loss_mlp": 1.02609074,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.9273352112365594,
      "language_loss": 0.79143786,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81343389,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 3.7686948776245117
    },
    {
      "auxiliary_loss_clip": 0.01184918,
      "auxiliary_loss_mlp": 0.01026333,
      "balance_loss_clip": 1.05798268,
      "balance_loss_mlp": 1.0182445,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.0579290163285813,
      "language_loss": 0.8091858,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.83129823,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.679603099822998
    },
    {
      "auxiliary_loss_clip": 0.0116363,
      "auxiliary_loss_mlp": 0.01036778,
      "balance_loss_clip": 1.05176795,
      "balance_loss_mlp": 1.02815938,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 3.403319617181145,
      "language_loss": 0.75917637,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.78118044,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.695533275604248
    },
    {
      "auxiliary_loss_clip": 0.0117725,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.05756855,
      "balance_loss_mlp": 1.02376246,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 2.0295107077239787,
      "language_loss": 0.80142462,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82350814,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 3.56184983253479
    },
    {
      "auxiliary_loss_clip": 0.01152897,
      "auxiliary_loss_mlp": 0.01032652,
      "balance_loss_clip": 1.0521853,
      "balance_loss_mlp": 1.02472448,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.0965242495124037,
      "language_loss": 0.8130399,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83489537,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.7622673511505127
    },
    {
      "auxiliary_loss_clip": 0.01170861,
      "auxiliary_loss_mlp": 0.01034055,
      "balance_loss_clip": 1.05243218,
      "balance_loss_mlp": 1.02674162,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.5476272691676995,
      "language_loss": 0.76823211,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.7902813,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.701470136642456
    },
    {
      "auxiliary_loss_clip": 0.01164284,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04915929,
      "balance_loss_mlp": 1.01658607,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.2467163444363085,
      "language_loss": 0.6726467,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.69453222,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 3.6695315837860107
    },
    {
      "auxiliary_loss_clip": 0.01191074,
      "auxiliary_loss_mlp": 0.010285,
      "balance_loss_clip": 1.06044853,
      "balance_loss_mlp": 1.02112067,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.1244667390438265,
      "language_loss": 0.84901392,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.87120962,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.538045644760132
    },
    {
      "auxiliary_loss_clip": 0.01187254,
      "auxiliary_loss_mlp": 0.01029018,
      "balance_loss_clip": 1.05580628,
      "balance_loss_mlp": 1.02095985,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 3.343844034945201,
      "language_loss": 0.66241026,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68457294,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.662745237350464
    },
    {
      "auxiliary_loss_clip": 0.01095561,
      "auxiliary_loss_mlp": 0.01001509,
      "balance_loss_clip": 1.02421975,
      "balance_loss_mlp": 1.00017953,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7378304531017927,
      "language_loss": 0.55471563,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57568634,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.126621723175049
    },
    {
      "auxiliary_loss_clip": 0.01081352,
      "auxiliary_loss_mlp": 0.01002296,
      "balance_loss_clip": 1.02068639,
      "balance_loss_mlp": 1.00103235,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7650459661660254,
      "language_loss": 0.61233228,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63316876,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1140897274017334
    },
    {
      "auxiliary_loss_clip": 0.01148058,
      "auxiliary_loss_mlp": 0.01027072,
      "balance_loss_clip": 1.05124068,
      "balance_loss_mlp": 1.01956439,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.4339238355845945,
      "language_loss": 0.82307625,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84482753,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.71600079536438
    },
    {
      "auxiliary_loss_clip": 0.01079327,
      "auxiliary_loss_mlp": 0.01000993,
      "balance_loss_clip": 1.03288364,
      "balance_loss_mlp": 0.99972337,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8856764777420237,
      "language_loss": 0.58734572,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60814893,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.179671049118042
    },
    {
      "auxiliary_loss_clip": 0.0118341,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.05860472,
      "balance_loss_mlp": 1.02466154,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.7940821773073627,
      "language_loss": 0.81057584,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.8327359,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.7640721797943115
    },
    {
      "auxiliary_loss_clip": 0.01170329,
      "auxiliary_loss_mlp": 0.01026506,
      "balance_loss_clip": 1.05632973,
      "balance_loss_mlp": 1.01861739,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 2.385807087681102,
      "language_loss": 0.78008199,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.80205035,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.6915805339813232
    },
    {
      "auxiliary_loss_clip": 0.01136075,
      "auxiliary_loss_mlp": 0.01023616,
      "balance_loss_clip": 1.04651833,
      "balance_loss_mlp": 1.01502657,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.6857947089499359,
      "language_loss": 0.81894183,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84053874,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.868079900741577
    },
    {
      "auxiliary_loss_clip": 0.01167322,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.05577922,
      "balance_loss_mlp": 1.02651513,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 2.8491068159744284,
      "language_loss": 0.82217324,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84419483,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.6905136108398438
    },
    {
      "auxiliary_loss_clip": 0.01184827,
      "auxiliary_loss_mlp": 0.01032202,
      "balance_loss_clip": 1.05711663,
      "balance_loss_mlp": 1.02422071,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 1.9723115647739395,
      "language_loss": 0.87732565,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.89949596,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.6068005561828613
    },
    {
      "auxiliary_loss_clip": 0.01158214,
      "auxiliary_loss_mlp": 0.00902587,
      "balance_loss_clip": 1.04875517,
      "balance_loss_mlp": 1.00155866,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 2.146652409354322,
      "language_loss": 0.85992301,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.88053101,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.714334487915039
    },
    {
      "auxiliary_loss_clip": 0.01178178,
      "auxiliary_loss_mlp": 0.00902098,
      "balance_loss_clip": 1.05195224,
      "balance_loss_mlp": 1.00156331,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 2.6660504723878358,
      "language_loss": 0.82715237,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.84795511,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.7348549365997314
    },
    {
      "auxiliary_loss_clip": 0.01171112,
      "auxiliary_loss_mlp": 0.01025223,
      "balance_loss_clip": 1.05751288,
      "balance_loss_mlp": 1.01726604,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 2.2876288909218276,
      "language_loss": 0.66024244,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.6822058,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.6356475353240967
    },
    {
      "auxiliary_loss_clip": 0.01180695,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.05511057,
      "balance_loss_mlp": 1.01666927,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.0208618958327316,
      "language_loss": 0.6527645,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67481625,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.683690071105957
    },
    {
      "auxiliary_loss_clip": 0.01184488,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.05536354,
      "balance_loss_mlp": 1.01783717,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.7223198349867073,
      "language_loss": 0.76437658,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78648114,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.732731580734253
    },
    {
      "auxiliary_loss_clip": 0.01196889,
      "auxiliary_loss_mlp": 0.01025803,
      "balance_loss_clip": 1.06029558,
      "balance_loss_mlp": 1.01766729,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 4.5528043107590905,
      "language_loss": 0.77367514,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79590207,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 3.685659646987915
    },
    {
      "auxiliary_loss_clip": 0.01149392,
      "auxiliary_loss_mlp": 0.01032209,
      "balance_loss_clip": 1.05043614,
      "balance_loss_mlp": 1.02412677,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.7572590120231468,
      "language_loss": 0.87108564,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89290166,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.771859645843506
    },
    {
      "auxiliary_loss_clip": 0.0115746,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.05135465,
      "balance_loss_mlp": 1.01859868,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.7117421923541842,
      "language_loss": 0.93101007,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95285058,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.6649744510650635
    },
    {
      "auxiliary_loss_clip": 0.0118067,
      "auxiliary_loss_mlp": 0.00901446,
      "balance_loss_clip": 1.05617893,
      "balance_loss_mlp": 1.00157523,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.5231691542526264,
      "language_loss": 0.65401971,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67484081,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.6468029022216797
    },
    {
      "auxiliary_loss_clip": 0.01159334,
      "auxiliary_loss_mlp": 0.00901902,
      "balance_loss_clip": 1.05135059,
      "balance_loss_mlp": 1.00159574,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.2796772250648667,
      "language_loss": 0.94567907,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96629149,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 3.6998555660247803
    },
    {
      "auxiliary_loss_clip": 0.01182136,
      "auxiliary_loss_mlp": 0.01026607,
      "balance_loss_clip": 1.0558095,
      "balance_loss_mlp": 1.01800621,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.685523197892804,
      "language_loss": 0.87676483,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.89885229,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.65043306350708
    },
    {
      "auxiliary_loss_clip": 0.01180578,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.05698466,
      "balance_loss_mlp": 1.02022696,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 6.830052213268889,
      "language_loss": 0.77612853,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79821616,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 3.602541446685791
    },
    {
      "auxiliary_loss_clip": 0.01171285,
      "auxiliary_loss_mlp": 0.01028145,
      "balance_loss_clip": 1.0520668,
      "balance_loss_mlp": 1.01984251,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.873553021817676,
      "language_loss": 0.85038882,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87238312,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.723104238510132
    },
    {
      "auxiliary_loss_clip": 0.01180901,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.0580554,
      "balance_loss_mlp": 1.02145934,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 3.810816571107145,
      "language_loss": 0.79897207,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82106954,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.754192590713501
    },
    {
      "auxiliary_loss_clip": 0.01171272,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.05116093,
      "balance_loss_mlp": 1.01817369,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 1.8912824250658535,
      "language_loss": 0.9034909,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92546821,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.757915735244751
    },
    {
      "auxiliary_loss_clip": 0.011519,
      "auxiliary_loss_mlp": 0.01033749,
      "balance_loss_clip": 1.04744828,
      "balance_loss_mlp": 1.02611995,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 3.1430319909786317,
      "language_loss": 0.75637686,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77823341,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 3.715893507003784
    },
    {
      "auxiliary_loss_clip": 0.01068319,
      "auxiliary_loss_mlp": 0.01007853,
      "balance_loss_clip": 1.01708388,
      "balance_loss_mlp": 1.00654817,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7878720653907644,
      "language_loss": 0.60855377,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.6293155,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.1077558994293213
    },
    {
      "auxiliary_loss_clip": 0.01182199,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.05383086,
      "balance_loss_mlp": 1.01753569,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 1.9049334662620552,
      "language_loss": 0.81188452,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83396626,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.6954452991485596
    },
    {
      "auxiliary_loss_clip": 0.01180963,
      "auxiliary_loss_mlp": 0.01033668,
      "balance_loss_clip": 1.05538428,
      "balance_loss_mlp": 1.02554095,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.9851164155589676,
      "language_loss": 0.68737859,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70952487,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.6789114475250244
    },
    {
      "auxiliary_loss_clip": 0.01118976,
      "auxiliary_loss_mlp": 0.01030167,
      "balance_loss_clip": 1.04523015,
      "balance_loss_mlp": 1.02156639,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 1.864795093540033,
      "language_loss": 0.80009127,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82158273,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.877830743789673
    },
    {
      "auxiliary_loss_clip": 0.01174198,
      "auxiliary_loss_mlp": 0.01029299,
      "balance_loss_clip": 1.05404282,
      "balance_loss_mlp": 1.02135348,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.7541110386398846,
      "language_loss": 0.80289847,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82493341,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.6989893913269043
    },
    {
      "auxiliary_loss_clip": 0.01186473,
      "auxiliary_loss_mlp": 0.01028081,
      "balance_loss_clip": 1.05465555,
      "balance_loss_mlp": 1.019611,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.6321290216583626,
      "language_loss": 0.87356281,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89570838,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.646195650100708
    },
    {
      "auxiliary_loss_clip": 0.01178313,
      "auxiliary_loss_mlp": 0.01035995,
      "balance_loss_clip": 1.0537703,
      "balance_loss_mlp": 1.02772832,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.9693916054713763,
      "language_loss": 0.82083398,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.84297705,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.594350576400757
    },
    {
      "auxiliary_loss_clip": 0.01160778,
      "auxiliary_loss_mlp": 0.01025897,
      "balance_loss_clip": 1.04913807,
      "balance_loss_mlp": 1.01804721,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.3017159062354535,
      "language_loss": 0.81316876,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83503556,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.8668837547302246
    },
    {
      "auxiliary_loss_clip": 0.01160838,
      "auxiliary_loss_mlp": 0.01025321,
      "balance_loss_clip": 1.05393958,
      "balance_loss_mlp": 1.01797175,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.7377190352303022,
      "language_loss": 0.66382611,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68568766,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.7217071056365967
    },
    {
      "auxiliary_loss_clip": 0.01187592,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.05388546,
      "balance_loss_mlp": 1.01609325,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.4806643101740256,
      "language_loss": 0.7545473,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.77666491,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.6265499591827393
    },
    {
      "auxiliary_loss_clip": 0.01154623,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.05050671,
      "balance_loss_mlp": 1.01929903,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 2.154161011888478,
      "language_loss": 0.88289535,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90471137,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.680638551712036
    },
    {
      "auxiliary_loss_clip": 0.01086277,
      "auxiliary_loss_mlp": 0.01001151,
      "balance_loss_clip": 1.02049255,
      "balance_loss_mlp": 0.99993461,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8142999626001518,
      "language_loss": 0.53304994,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.5539242,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.227076768875122
    },
    {
      "auxiliary_loss_clip": 0.01168048,
      "auxiliary_loss_mlp": 0.01029728,
      "balance_loss_clip": 1.0539608,
      "balance_loss_mlp": 1.02137101,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.4178317371968934,
      "language_loss": 0.69945574,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.72143352,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.6360840797424316
    },
    {
      "auxiliary_loss_clip": 0.0116735,
      "auxiliary_loss_mlp": 0.0090136,
      "balance_loss_clip": 1.05296254,
      "balance_loss_mlp": 1.00158572,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 3.362527399652751,
      "language_loss": 0.66005927,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68074632,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.743265390396118
    },
    {
      "auxiliary_loss_clip": 0.01142698,
      "auxiliary_loss_mlp": 0.01037608,
      "balance_loss_clip": 1.04930389,
      "balance_loss_mlp": 1.02864289,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 3.3873104432237624,
      "language_loss": 0.6445266,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.66632962,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.812990188598633
    },
    {
      "auxiliary_loss_clip": 0.0116368,
      "auxiliary_loss_mlp": 0.01028241,
      "balance_loss_clip": 1.05220604,
      "balance_loss_mlp": 1.01995575,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 2.9318799832579847,
      "language_loss": 0.76210749,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78402674,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 3.6486523151397705
    },
    {
      "auxiliary_loss_clip": 0.01156628,
      "auxiliary_loss_mlp": 0.01032234,
      "balance_loss_clip": 1.04758584,
      "balance_loss_mlp": 1.02354991,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.359827228693361,
      "language_loss": 0.70573747,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72762614,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.8195669651031494
    },
    {
      "auxiliary_loss_clip": 0.01180943,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.0559119,
      "balance_loss_mlp": 1.02465141,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 2.17962961123666,
      "language_loss": 0.84070802,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86285162,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.682055711746216
    },
    {
      "auxiliary_loss_clip": 0.01189644,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 1.05700588,
      "balance_loss_mlp": 1.0190084,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 2.373936657965404,
      "language_loss": 0.75442535,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77658731,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.692876100540161
    },
    {
      "auxiliary_loss_clip": 0.01179057,
      "auxiliary_loss_mlp": 0.01040763,
      "balance_loss_clip": 1.05500531,
      "balance_loss_mlp": 1.03203082,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 2.7951641247690255,
      "language_loss": 0.7722863,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79448456,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 3.568526029586792
    },
    {
      "auxiliary_loss_clip": 0.01163236,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.0523411,
      "balance_loss_mlp": 1.01715899,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.9409057769145686,
      "language_loss": 0.8180604,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.83995444,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.678678035736084
    },
    {
      "auxiliary_loss_clip": 0.01175859,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.05543995,
      "balance_loss_mlp": 1.01755416,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 1.8296134631425467,
      "language_loss": 0.88497555,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90698957,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.722344398498535
    },
    {
      "auxiliary_loss_clip": 0.01177052,
      "auxiliary_loss_mlp": 0.01031703,
      "balance_loss_clip": 1.05351686,
      "balance_loss_mlp": 1.02375746,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.757474482754682,
      "language_loss": 0.77116221,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79324973,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 3.6891515254974365
    },
    {
      "auxiliary_loss_clip": 0.01183813,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 1.05625868,
      "balance_loss_mlp": 1.01806104,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.8388009355481714,
      "language_loss": 0.65674418,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67883873,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.6770176887512207
    },
    {
      "auxiliary_loss_clip": 0.01188228,
      "auxiliary_loss_mlp": 0.01027882,
      "balance_loss_clip": 1.05666971,
      "balance_loss_mlp": 1.01971602,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.3949122674251204,
      "language_loss": 0.87101352,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89317465,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.6690165996551514
    },
    {
      "auxiliary_loss_clip": 0.01191827,
      "auxiliary_loss_mlp": 0.01025257,
      "balance_loss_clip": 1.05746078,
      "balance_loss_mlp": 1.01722836,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 2.3199730445313906,
      "language_loss": 0.78787267,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81004351,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.551363468170166
    },
    {
      "auxiliary_loss_clip": 0.01141833,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 1.04763317,
      "balance_loss_mlp": 1.02058649,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 11.30238110205832,
      "language_loss": 0.67790687,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.69961739,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 3.6875314712524414
    },
    {
      "auxiliary_loss_clip": 0.01163078,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05202341,
      "balance_loss_mlp": 1.017452,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.1969111442494085,
      "language_loss": 0.63848776,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66038042,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.6296534538269043
    },
    {
      "auxiliary_loss_clip": 0.01191359,
      "auxiliary_loss_mlp": 0.01036519,
      "balance_loss_clip": 1.05691051,
      "balance_loss_mlp": 1.02882433,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 22.932846824231287,
      "language_loss": 0.8393842,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.86166292,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.629394769668579
    },
    {
      "auxiliary_loss_clip": 0.0117462,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 1.05290508,
      "balance_loss_mlp": 1.01730645,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 3.001751105697302,
      "language_loss": 0.75709713,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77909684,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.868767023086548
    },
    {
      "auxiliary_loss_clip": 0.01191041,
      "auxiliary_loss_mlp": 0.01029845,
      "balance_loss_clip": 1.05739355,
      "balance_loss_mlp": 1.02188182,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.8598170473692626,
      "language_loss": 0.73252702,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75473589,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.662742853164673
    },
    {
      "auxiliary_loss_clip": 0.0119455,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.0591495,
      "balance_loss_mlp": 1.01966333,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 3.5587904697404453,
      "language_loss": 0.79216623,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81439084,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.6108689308166504
    },
    {
      "auxiliary_loss_clip": 0.01183441,
      "auxiliary_loss_mlp": 0.00902107,
      "balance_loss_clip": 1.05624247,
      "balance_loss_mlp": 1.0017035,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.118845758806057,
      "language_loss": 0.76538253,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.78623801,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.6387219429016113
    },
    {
      "auxiliary_loss_clip": 0.01175195,
      "auxiliary_loss_mlp": 0.01032775,
      "balance_loss_clip": 1.05182314,
      "balance_loss_mlp": 1.0242275,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 2.815964859970755,
      "language_loss": 0.89626813,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.91834784,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.6383519172668457
    },
    {
      "auxiliary_loss_clip": 0.01077998,
      "auxiliary_loss_mlp": 0.01001052,
      "balance_loss_clip": 1.01852572,
      "balance_loss_mlp": 0.99991339,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7650775812472481,
      "language_loss": 0.57347298,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59426349,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.1809067726135254
    },
    {
      "auxiliary_loss_clip": 0.01162552,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04974973,
      "balance_loss_mlp": 1.01679635,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.9905471686718887,
      "language_loss": 0.78518796,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80705905,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.7118301391601562
    },
    {
      "auxiliary_loss_clip": 0.01142792,
      "auxiliary_loss_mlp": 0.01032127,
      "balance_loss_clip": 1.04920304,
      "balance_loss_mlp": 1.02396703,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 2.1159850073249893,
      "language_loss": 0.86504853,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88679767,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.8132717609405518
    },
    {
      "auxiliary_loss_clip": 0.01181862,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.05242801,
      "balance_loss_mlp": 1.01663947,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 1.9476425250911211,
      "language_loss": 0.7506038,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77267349,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.787844657897949
    },
    {
      "auxiliary_loss_clip": 0.0114529,
      "auxiliary_loss_mlp": 0.01029993,
      "balance_loss_clip": 1.04571366,
      "balance_loss_mlp": 1.02113616,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.3922381003834623,
      "language_loss": 0.72204238,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74379516,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.652250051498413
    },
    {
      "auxiliary_loss_clip": 0.01164257,
      "auxiliary_loss_mlp": 0.01031816,
      "balance_loss_clip": 1.05447054,
      "balance_loss_mlp": 1.02370405,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.6799350900634,
      "language_loss": 0.61335987,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.6353206,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.7020888328552246
    },
    {
      "auxiliary_loss_clip": 0.01170485,
      "auxiliary_loss_mlp": 0.01024841,
      "balance_loss_clip": 1.05715489,
      "balance_loss_mlp": 1.01632357,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 2.5199954960438613,
      "language_loss": 0.74756694,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.76952016,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.6298165321350098
    },
    {
      "auxiliary_loss_clip": 0.01186511,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.05670226,
      "balance_loss_mlp": 1.02134538,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.8780207218614065,
      "language_loss": 0.80999994,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83216274,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 3.566286325454712
    },
    {
      "auxiliary_loss_clip": 0.01145563,
      "auxiliary_loss_mlp": 0.01029861,
      "balance_loss_clip": 1.04580522,
      "balance_loss_mlp": 1.02186155,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 1.993960385247558,
      "language_loss": 0.82371652,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84547073,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.712773561477661
    },
    {
      "auxiliary_loss_clip": 0.01163394,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.0510596,
      "balance_loss_mlp": 1.02194703,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 2.0024366045835267,
      "language_loss": 0.61645114,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63838696,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.7178614139556885
    },
    {
      "auxiliary_loss_clip": 0.01154356,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.05164552,
      "balance_loss_mlp": 1.02285135,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 1.928129200520063,
      "language_loss": 0.87626684,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89812875,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.767653703689575
    },
    {
      "auxiliary_loss_clip": 0.01164548,
      "auxiliary_loss_mlp": 0.00902095,
      "balance_loss_clip": 1.05258083,
      "balance_loss_mlp": 1.00158882,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.4674863072641884,
      "language_loss": 0.75159734,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.77226377,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 3.6058287620544434
    },
    {
      "auxiliary_loss_clip": 0.0115784,
      "auxiliary_loss_mlp": 0.01031161,
      "balance_loss_clip": 1.04693091,
      "balance_loss_mlp": 1.02260733,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.681656052546333,
      "language_loss": 0.84170157,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86359155,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.6976821422576904
    },
    {
      "auxiliary_loss_clip": 0.01175663,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.05269003,
      "balance_loss_mlp": 1.01911664,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 1.9230304937392737,
      "language_loss": 0.72957587,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75160599,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.7396063804626465
    },
    {
      "auxiliary_loss_clip": 0.01189627,
      "auxiliary_loss_mlp": 0.0103155,
      "balance_loss_clip": 1.05611193,
      "balance_loss_mlp": 1.02286005,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 2.006138479977237,
      "language_loss": 0.7321381,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75434983,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 3.5358169078826904
    },
    {
      "auxiliary_loss_clip": 0.0115744,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 1.050143,
      "balance_loss_mlp": 1.01808858,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 1.926832566697075,
      "language_loss": 0.84893274,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.87076443,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.737661600112915
    },
    {
      "auxiliary_loss_clip": 0.01185203,
      "auxiliary_loss_mlp": 0.01025163,
      "balance_loss_clip": 1.05413103,
      "balance_loss_mlp": 1.01692522,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 1.8934353246199256,
      "language_loss": 0.80598426,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82808793,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.654317855834961
    },
    {
      "auxiliary_loss_clip": 0.01177314,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.05287766,
      "balance_loss_mlp": 1.02089751,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.8117926219114975,
      "language_loss": 0.79046792,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81253028,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.7181167602539062
    },
    {
      "auxiliary_loss_clip": 0.01177059,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.05244493,
      "balance_loss_mlp": 1.02460694,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.791284389791887,
      "language_loss": 0.7743659,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.7964679,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 3.5281050205230713
    },
    {
      "auxiliary_loss_clip": 0.01192228,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.05575216,
      "balance_loss_mlp": 1.01964736,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.9869900264107188,
      "language_loss": 0.79520464,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81741321,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.667825698852539
    },
    {
      "auxiliary_loss_clip": 0.01179236,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.05642128,
      "balance_loss_mlp": 1.02165627,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.023299989795542,
      "language_loss": 0.8451637,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.86725003,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.628976583480835
    },
    {
      "auxiliary_loss_clip": 0.01178562,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.05585432,
      "balance_loss_mlp": 1.01722622,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.6389835912647792,
      "language_loss": 0.80411118,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82614815,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.6710426807403564
    },
    {
      "auxiliary_loss_clip": 0.0118376,
      "auxiliary_loss_mlp": 0.01024397,
      "balance_loss_clip": 1.05621314,
      "balance_loss_mlp": 1.01594472,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.6204522743109413,
      "language_loss": 0.76562095,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.7877025,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.6532444953918457
    },
    {
      "auxiliary_loss_clip": 0.01164575,
      "auxiliary_loss_mlp": 0.01029444,
      "balance_loss_clip": 1.05084014,
      "balance_loss_mlp": 1.0207653,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 1.9977814108347023,
      "language_loss": 0.73274642,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.75468659,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.671084403991699
    },
    {
      "auxiliary_loss_clip": 0.01178125,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.05262673,
      "balance_loss_mlp": 1.02551877,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 2.631008884254925,
      "language_loss": 0.75487113,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77698731,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.685128927230835
    },
    {
      "auxiliary_loss_clip": 0.01082307,
      "auxiliary_loss_mlp": 0.01003024,
      "balance_loss_clip": 1.02116227,
      "balance_loss_mlp": 1.00186121,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7574251605295683,
      "language_loss": 0.56517172,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.586025,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.097288131713867
    },
    {
      "auxiliary_loss_clip": 0.01180683,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.05541587,
      "balance_loss_mlp": 1.01651621,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 2.706671473478047,
      "language_loss": 0.69790196,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.71995497,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.6151680946350098
    },
    {
      "auxiliary_loss_clip": 0.01165038,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.05580354,
      "balance_loss_mlp": 1.02030802,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.5112956454850868,
      "language_loss": 0.73927796,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.76121485,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.7023959159851074
    },
    {
      "auxiliary_loss_clip": 0.01183018,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.05437171,
      "balance_loss_mlp": 1.02162588,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 1.9928142290615594,
      "language_loss": 0.78919113,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.81132197,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.540342330932617
    },
    {
      "auxiliary_loss_clip": 0.01175052,
      "auxiliary_loss_mlp": 0.01028295,
      "balance_loss_clip": 1.05388975,
      "balance_loss_mlp": 1.02018881,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.519608148365422,
      "language_loss": 0.82756501,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.84959847,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.6727023124694824
    },
    {
      "auxiliary_loss_clip": 0.01166285,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.05144763,
      "balance_loss_mlp": 1.01646292,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.8703888209984818,
      "language_loss": 0.80892754,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83083451,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.6317224502563477
    },
    {
      "auxiliary_loss_clip": 0.01197085,
      "auxiliary_loss_mlp": 0.0090241,
      "balance_loss_clip": 1.05826306,
      "balance_loss_mlp": 1.00154567,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 2.793214722504188,
      "language_loss": 0.77487409,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.79586899,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.5594584941864014
    },
    {
      "auxiliary_loss_clip": 0.01150305,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.0541327,
      "balance_loss_mlp": 1.01717961,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 4.363211155428089,
      "language_loss": 0.74072152,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76247716,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.6826491355895996
    },
    {
      "auxiliary_loss_clip": 0.01185942,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.05819452,
      "balance_loss_mlp": 1.02449882,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 5.472226364347682,
      "language_loss": 0.68725538,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70944512,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 2.710568428039551
    },
    {
      "auxiliary_loss_clip": 0.01174912,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.05255389,
      "balance_loss_mlp": 1.02146912,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 3.4377716605150375,
      "language_loss": 0.6478768,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66992533,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 3.6435370445251465
    },
    {
      "auxiliary_loss_clip": 0.0115706,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.05332851,
      "balance_loss_mlp": 1.0197562,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.73317725645329,
      "language_loss": 0.71766162,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.73951048,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 2.7294766902923584
    },
    {
      "auxiliary_loss_clip": 0.01146082,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.05165648,
      "balance_loss_mlp": 1.02354813,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.6330395825944306,
      "language_loss": 0.66287071,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68465304,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.7801594734191895
    },
    {
      "auxiliary_loss_clip": 0.01192257,
      "auxiliary_loss_mlp": 0.01030399,
      "balance_loss_clip": 1.05673075,
      "balance_loss_mlp": 1.02208996,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 3.15916175199177,
      "language_loss": 0.81012809,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.83235466,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.5817792415618896
    },
    {
      "auxiliary_loss_clip": 0.01164955,
      "auxiliary_loss_mlp": 0.01028013,
      "balance_loss_clip": 1.05112171,
      "balance_loss_mlp": 1.01947212,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.0549701985333106,
      "language_loss": 0.82101107,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84294075,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.63285231590271
    },
    {
      "auxiliary_loss_clip": 0.01192533,
      "auxiliary_loss_mlp": 0.01027147,
      "balance_loss_clip": 1.05768251,
      "balance_loss_mlp": 1.01923156,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.614191266682157,
      "language_loss": 0.82863581,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85083258,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 3.4120094776153564
    },
    {
      "auxiliary_loss_clip": 0.01163793,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.04957938,
      "balance_loss_mlp": 1.02321005,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.7876036484923987,
      "language_loss": 0.82100338,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.84296298,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.652836799621582
    },
    {
      "auxiliary_loss_clip": 0.01163289,
      "auxiliary_loss_mlp": 0.01025762,
      "balance_loss_clip": 1.05144203,
      "balance_loss_mlp": 1.01705968,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 1.975421152253347,
      "language_loss": 0.89975643,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.92164695,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 3.579984426498413
    },
    {
      "auxiliary_loss_clip": 0.0119381,
      "auxiliary_loss_mlp": 0.01039842,
      "balance_loss_clip": 1.05774379,
      "balance_loss_mlp": 1.0311873,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.167388951484914,
      "language_loss": 0.81113958,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83347607,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.6109583377838135
    },
    {
      "auxiliary_loss_clip": 0.01182105,
      "auxiliary_loss_mlp": 0.01033975,
      "balance_loss_clip": 1.05504751,
      "balance_loss_mlp": 1.02518952,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 2.244804810016241,
      "language_loss": 0.64072388,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66288471,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.6740736961364746
    },
    {
      "auxiliary_loss_clip": 0.01179639,
      "auxiliary_loss_mlp": 0.01030367,
      "balance_loss_clip": 1.05262876,
      "balance_loss_mlp": 1.02129507,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.637437518497063,
      "language_loss": 0.60591424,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62801433,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.622532367706299
    },
    {
      "auxiliary_loss_clip": 0.01162564,
      "auxiliary_loss_mlp": 0.01033212,
      "balance_loss_clip": 1.04973006,
      "balance_loss_mlp": 1.02498686,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 2.611630446986158,
      "language_loss": 0.75954789,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.7815057,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 3.6842660903930664
    },
    {
      "auxiliary_loss_clip": 0.01180506,
      "auxiliary_loss_mlp": 0.01028679,
      "balance_loss_clip": 1.05766869,
      "balance_loss_mlp": 1.02076352,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.8495522460411045,
      "language_loss": 0.71517098,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73726279,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.6492526531219482
    },
    {
      "auxiliary_loss_clip": 0.01071319,
      "auxiliary_loss_mlp": 0.01000565,
      "balance_loss_clip": 1.02034998,
      "balance_loss_mlp": 0.99937338,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8459384427517159,
      "language_loss": 0.59549117,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61621004,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.3641419410705566
    },
    {
      "auxiliary_loss_clip": 0.0117945,
      "auxiliary_loss_mlp": 0.01031032,
      "balance_loss_clip": 1.05221176,
      "balance_loss_mlp": 1.02295017,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.52359060310306,
      "language_loss": 0.79604715,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81815201,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.738459587097168
    },
    {
      "auxiliary_loss_clip": 0.01159002,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.0522511,
      "balance_loss_mlp": 1.02016985,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.3545641424633117,
      "language_loss": 0.74725616,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.76912081,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.608546495437622
    },
    {
      "auxiliary_loss_clip": 0.01144532,
      "auxiliary_loss_mlp": 0.01025242,
      "balance_loss_clip": 1.0497824,
      "balance_loss_mlp": 1.01748109,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.6560083807343913,
      "language_loss": 0.684358,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70605576,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.812650442123413
    },
    {
      "auxiliary_loss_clip": 0.01181735,
      "auxiliary_loss_mlp": 0.01032852,
      "balance_loss_clip": 1.05954695,
      "balance_loss_mlp": 1.02451921,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 1.9212350273165273,
      "language_loss": 0.71196675,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73411262,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.5939345359802246
    },
    {
      "auxiliary_loss_clip": 0.01165943,
      "auxiliary_loss_mlp": 0.01031375,
      "balance_loss_clip": 1.05133986,
      "balance_loss_mlp": 1.02269626,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 2.1907023947298523,
      "language_loss": 0.83573371,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85770684,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.642570972442627
    },
    {
      "auxiliary_loss_clip": 0.01184006,
      "auxiliary_loss_mlp": 0.00902354,
      "balance_loss_clip": 1.05866456,
      "balance_loss_mlp": 1.00166917,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.6950956792389604,
      "language_loss": 0.65592551,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.6767891,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.6305091381073
    },
    {
      "auxiliary_loss_clip": 0.01190132,
      "auxiliary_loss_mlp": 0.01026459,
      "balance_loss_clip": 1.05714202,
      "balance_loss_mlp": 1.01855564,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 1.9268660984461345,
      "language_loss": 0.72502685,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.74719274,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.6032421588897705
    },
    {
      "auxiliary_loss_clip": 0.01160179,
      "auxiliary_loss_mlp": 0.01028533,
      "balance_loss_clip": 1.05415046,
      "balance_loss_mlp": 1.02016497,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 1.9535066090028828,
      "language_loss": 0.77827621,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.80016333,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.6055967807769775
    },
    {
      "auxiliary_loss_clip": 0.01173058,
      "auxiliary_loss_mlp": 0.01028306,
      "balance_loss_clip": 1.05277717,
      "balance_loss_mlp": 1.0202055,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 2.0867453057444454,
      "language_loss": 0.79559237,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81760603,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.6189091205596924
    },
    {
      "auxiliary_loss_clip": 0.01149947,
      "auxiliary_loss_mlp": 0.01032417,
      "balance_loss_clip": 1.04945445,
      "balance_loss_mlp": 1.0241735,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.9246943804127998,
      "language_loss": 0.7040906,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72591424,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.6932373046875
    },
    {
      "auxiliary_loss_clip": 0.01191374,
      "auxiliary_loss_mlp": 0.01031963,
      "balance_loss_clip": 1.05866528,
      "balance_loss_mlp": 1.02408957,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.0176251800147855,
      "language_loss": 0.76557374,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78780711,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.572568893432617
    },
    {
      "auxiliary_loss_clip": 0.0118705,
      "auxiliary_loss_mlp": 0.01025612,
      "balance_loss_clip": 1.05365372,
      "balance_loss_mlp": 1.01723754,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 2.22715242075083,
      "language_loss": 0.78610313,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80822974,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.573667526245117
    },
    {
      "auxiliary_loss_clip": 0.01149726,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.04772627,
      "balance_loss_mlp": 1.02232766,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.920902541752487,
      "language_loss": 0.82937717,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.85119027,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 3.6889736652374268
    },
    {
      "auxiliary_loss_clip": 0.01176317,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.05174255,
      "balance_loss_mlp": 1.01785994,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 1.9055218692474012,
      "language_loss": 0.85374069,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87576842,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.620476007461548
    },
    {
      "auxiliary_loss_clip": 0.01170683,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.05171323,
      "balance_loss_mlp": 1.01735771,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 2.112207434179259,
      "language_loss": 0.71530521,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73726708,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.723602294921875
    },
    {
      "auxiliary_loss_clip": 0.01185878,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05220878,
      "balance_loss_mlp": 1.02405679,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.8824510981052256,
      "language_loss": 0.7767309,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79891628,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6432251930236816
    },
    {
      "auxiliary_loss_clip": 0.01173077,
      "auxiliary_loss_mlp": 0.01022047,
      "balance_loss_clip": 1.05213642,
      "balance_loss_mlp": 1.01468015,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.6949345385844976,
      "language_loss": 0.81197488,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.8339262,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.619088888168335
    },
    {
      "auxiliary_loss_clip": 0.01184832,
      "auxiliary_loss_mlp": 0.01029828,
      "balance_loss_clip": 1.05631816,
      "balance_loss_mlp": 1.02162671,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 4.485737811768373,
      "language_loss": 0.8075453,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.82969189,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.6655678749084473
    },
    {
      "auxiliary_loss_clip": 0.0118237,
      "auxiliary_loss_mlp": 0.01032792,
      "balance_loss_clip": 1.05530858,
      "balance_loss_mlp": 1.02457857,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 2.520574676042586,
      "language_loss": 0.8732909,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89544255,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 3.47087025642395
    },
    {
      "auxiliary_loss_clip": 0.0117319,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.05117202,
      "balance_loss_mlp": 1.01822472,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.7355314588586885,
      "language_loss": 0.79339576,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81538731,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 3.585110902786255
    },
    {
      "auxiliary_loss_clip": 0.01166021,
      "auxiliary_loss_mlp": 0.01029359,
      "balance_loss_clip": 1.04986787,
      "balance_loss_mlp": 1.02034044,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.5995223780277772,
      "language_loss": 0.82584977,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.84780359,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.693798065185547
    },
    {
      "auxiliary_loss_clip": 0.01160321,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.05035567,
      "balance_loss_mlp": 1.01795232,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.9742404064408228,
      "language_loss": 0.76851296,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.79037249,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.675204277038574
    },
    {
      "auxiliary_loss_clip": 0.01184323,
      "auxiliary_loss_mlp": 0.01027069,
      "balance_loss_clip": 1.05537844,
      "balance_loss_mlp": 1.01879549,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.367108966008161,
      "language_loss": 0.89920753,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92132139,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.6091482639312744
    },
    {
      "auxiliary_loss_clip": 0.01169091,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.05376267,
      "balance_loss_mlp": 1.01991749,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.632581212720812,
      "language_loss": 0.76650602,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.7884773,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 3.5966835021972656
    },
    {
      "auxiliary_loss_clip": 0.01173854,
      "auxiliary_loss_mlp": 0.01031386,
      "balance_loss_clip": 1.04984474,
      "balance_loss_mlp": 1.02326787,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 2.396466071856843,
      "language_loss": 0.81875086,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84080327,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.715822219848633
    },
    {
      "auxiliary_loss_clip": 0.01176804,
      "auxiliary_loss_mlp": 0.01026059,
      "balance_loss_clip": 1.05378151,
      "balance_loss_mlp": 1.01850128,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.6241388709148905,
      "language_loss": 0.6645661,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68659472,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.6295621395111084
    },
    {
      "auxiliary_loss_clip": 0.01187419,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.0545212,
      "balance_loss_mlp": 1.01812673,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.2187801395791906,
      "language_loss": 0.72705442,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74919534,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.741398811340332
    },
    {
      "auxiliary_loss_clip": 0.01159347,
      "auxiliary_loss_mlp": 0.01026961,
      "balance_loss_clip": 1.04950547,
      "balance_loss_mlp": 1.01886082,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.2261634507756045,
      "language_loss": 0.8910355,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.9128986,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.6532046794891357
    },
    {
      "auxiliary_loss_clip": 0.01076625,
      "auxiliary_loss_mlp": 0.01007519,
      "balance_loss_clip": 1.0185833,
      "balance_loss_mlp": 1.00629079,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6360804756576532,
      "language_loss": 0.48564112,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50648254,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.2717697620391846
    },
    {
      "auxiliary_loss_clip": 0.01155329,
      "auxiliary_loss_mlp": 0.01037353,
      "balance_loss_clip": 1.05327857,
      "balance_loss_mlp": 1.02917492,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.0997439117774803,
      "language_loss": 0.8995989,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.92152572,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.7199628353118896
    },
    {
      "auxiliary_loss_clip": 0.01182059,
      "auxiliary_loss_mlp": 0.0090212,
      "balance_loss_clip": 1.05519247,
      "balance_loss_mlp": 1.00164557,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 2.207361872669185,
      "language_loss": 0.86935449,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89019626,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.6047556400299072
    },
    {
      "auxiliary_loss_clip": 0.01170135,
      "auxiliary_loss_mlp": 0.01028047,
      "balance_loss_clip": 1.05145097,
      "balance_loss_mlp": 1.01962447,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.235324372295302,
      "language_loss": 0.66234583,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.6843276,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.6667168140411377
    },
    {
      "auxiliary_loss_clip": 0.01186096,
      "auxiliary_loss_mlp": 0.01022579,
      "balance_loss_clip": 1.05522418,
      "balance_loss_mlp": 1.01438951,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.8131292887877484,
      "language_loss": 0.81107336,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.8331601,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.6510329246520996
    },
    {
      "auxiliary_loss_clip": 0.01079591,
      "auxiliary_loss_mlp": 0.01003982,
      "balance_loss_clip": 1.02217865,
      "balance_loss_mlp": 1.0028677,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7719133681099213,
      "language_loss": 0.60450041,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62533611,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.262669324874878
    },
    {
      "auxiliary_loss_clip": 0.01177384,
      "auxiliary_loss_mlp": 0.01030938,
      "balance_loss_clip": 1.05097771,
      "balance_loss_mlp": 1.02337444,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 2.631063540949687,
      "language_loss": 0.6848371,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70692039,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.6634814739227295
    },
    {
      "auxiliary_loss_clip": 0.01169757,
      "auxiliary_loss_mlp": 0.01030909,
      "balance_loss_clip": 1.05209041,
      "balance_loss_mlp": 1.0230298,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.1714666900552313,
      "language_loss": 0.81450117,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.83650786,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.580239772796631
    },
    {
      "auxiliary_loss_clip": 0.01183832,
      "auxiliary_loss_mlp": 0.01029858,
      "balance_loss_clip": 1.05470228,
      "balance_loss_mlp": 1.02227628,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.3039603727009554,
      "language_loss": 0.83349985,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85563672,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.6546289920806885
    },
    {
      "auxiliary_loss_clip": 0.01181535,
      "auxiliary_loss_mlp": 0.01024866,
      "balance_loss_clip": 1.0539372,
      "balance_loss_mlp": 1.0173316,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 3.8097942998697403,
      "language_loss": 0.84836692,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87043095,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.6085643768310547
    },
    {
      "auxiliary_loss_clip": 0.01159555,
      "auxiliary_loss_mlp": 0.01028979,
      "balance_loss_clip": 1.0508281,
      "balance_loss_mlp": 1.02080107,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 2.4750906993989523,
      "language_loss": 0.61859834,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64048368,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 3.619041681289673
    },
    {
      "auxiliary_loss_clip": 0.01151845,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.04788578,
      "balance_loss_mlp": 1.02081776,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.6326479762834232,
      "language_loss": 0.83925343,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86105692,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 2.6812028884887695
    },
    {
      "auxiliary_loss_clip": 0.011823,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.05419374,
      "balance_loss_mlp": 1.02249861,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.8414559061778264,
      "language_loss": 0.7508406,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77296478,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.6238157749176025
    },
    {
      "auxiliary_loss_clip": 0.01172108,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.05317974,
      "balance_loss_mlp": 1.01753271,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.4064780942806692,
      "language_loss": 0.87703019,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89899933,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.6175131797790527
    },
    {
      "auxiliary_loss_clip": 0.011754,
      "auxiliary_loss_mlp": 0.01029741,
      "balance_loss_clip": 1.054003,
      "balance_loss_mlp": 1.02124178,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 2.5097519618234823,
      "language_loss": 0.76639563,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78844702,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.6245815753936768
    },
    {
      "auxiliary_loss_clip": 0.01185613,
      "auxiliary_loss_mlp": 0.01032192,
      "balance_loss_clip": 1.05627787,
      "balance_loss_mlp": 1.02403831,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.4198135864436248,
      "language_loss": 0.78587782,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80805582,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.6130282878875732
    },
    {
      "auxiliary_loss_clip": 0.01175446,
      "auxiliary_loss_mlp": 0.01031716,
      "balance_loss_clip": 1.05044007,
      "balance_loss_mlp": 1.0241431,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.7533697032567144,
      "language_loss": 0.82022882,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84230042,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 3.4902079105377197
    },
    {
      "auxiliary_loss_clip": 0.0115759,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.04881763,
      "balance_loss_mlp": 1.02000546,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 2.3682547386491515,
      "language_loss": 0.85867512,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.88053018,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.7117726802825928
    },
    {
      "auxiliary_loss_clip": 0.01148047,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.04946613,
      "balance_loss_mlp": 1.02016115,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.236653044092833,
      "language_loss": 0.74698198,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76874268,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 3.499216318130493
    },
    {
      "auxiliary_loss_clip": 0.01176536,
      "auxiliary_loss_mlp": 0.01032766,
      "balance_loss_clip": 1.05166054,
      "balance_loss_mlp": 1.0242486,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.794834999158086,
      "language_loss": 0.81936759,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84146059,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.610786199569702
    },
    {
      "auxiliary_loss_clip": 0.01155906,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.04870796,
      "balance_loss_mlp": 1.01746321,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 3.436861904983495,
      "language_loss": 0.74538475,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.76720357,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.7113001346588135
    },
    {
      "auxiliary_loss_clip": 0.01173254,
      "auxiliary_loss_mlp": 0.01033202,
      "balance_loss_clip": 1.04885662,
      "balance_loss_mlp": 1.02445197,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.298376575293555,
      "language_loss": 0.92976081,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95182532,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.6826791763305664
    },
    {
      "auxiliary_loss_clip": 0.01183542,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.05144477,
      "balance_loss_mlp": 1.01961815,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.3167284405416413,
      "language_loss": 0.91404724,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93615657,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 3.5858728885650635
    },
    {
      "auxiliary_loss_clip": 0.01169589,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.0483917,
      "balance_loss_mlp": 1.02038598,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.6788752850526198,
      "language_loss": 0.6775099,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69949275,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.6174404621124268
    },
    {
      "auxiliary_loss_clip": 0.01176266,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05001783,
      "balance_loss_mlp": 1.02615488,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 1.850559821629928,
      "language_loss": 0.7723223,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79442418,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.6942834854125977
    },
    {
      "auxiliary_loss_clip": 0.01162961,
      "auxiliary_loss_mlp": 0.01036775,
      "balance_loss_clip": 1.05104995,
      "balance_loss_mlp": 1.02844214,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.751806127326875,
      "language_loss": 0.72218859,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74418598,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.6656696796417236
    },
    {
      "auxiliary_loss_clip": 0.01178662,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.05368423,
      "balance_loss_mlp": 1.02082372,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 3.5984112216232385,
      "language_loss": 0.82945931,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.85154247,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.646108388900757
    },
    {
      "auxiliary_loss_clip": 0.01188794,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.05538619,
      "balance_loss_mlp": 1.02203798,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 2.36874009721202,
      "language_loss": 0.76168597,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.78387666,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.517188787460327
    },
    {
      "auxiliary_loss_clip": 0.01158554,
      "auxiliary_loss_mlp": 0.01031789,
      "balance_loss_clip": 1.04906261,
      "balance_loss_mlp": 1.02410626,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.7449951915417332,
      "language_loss": 0.66698194,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68888533,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.84800386428833
    },
    {
      "auxiliary_loss_clip": 0.01160854,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.04828525,
      "balance_loss_mlp": 1.01898026,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.7968196387682525,
      "language_loss": 0.65967357,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.68155807,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.641233444213867
    },
    {
      "auxiliary_loss_clip": 0.01183808,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.05329394,
      "balance_loss_mlp": 1.01848924,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.851386584469356,
      "language_loss": 0.70851135,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.73061401,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.6758625507354736
    },
    {
      "auxiliary_loss_clip": 0.0117884,
      "auxiliary_loss_mlp": 0.01030488,
      "balance_loss_clip": 1.05387294,
      "balance_loss_mlp": 1.02257276,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.523242284248884,
      "language_loss": 0.76679289,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78888619,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.602999448776245
    },
    {
      "auxiliary_loss_clip": 0.01163501,
      "auxiliary_loss_mlp": 0.01026394,
      "balance_loss_clip": 1.05523264,
      "balance_loss_mlp": 1.0193578,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.54216563242892,
      "language_loss": 0.79580402,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81770295,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.614745855331421
    },
    {
      "auxiliary_loss_clip": 0.01182552,
      "auxiliary_loss_mlp": 0.00901803,
      "balance_loss_clip": 1.05195665,
      "balance_loss_mlp": 1.00146842,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.326808967286902,
      "language_loss": 0.77355665,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79440022,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.6247923374176025
    },
    {
      "auxiliary_loss_clip": 0.01187269,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.0548346,
      "balance_loss_mlp": 1.01721215,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.0758978695854085,
      "language_loss": 0.87179518,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.89391935,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.5412299633026123
    },
    {
      "auxiliary_loss_clip": 0.01177949,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.0521183,
      "balance_loss_mlp": 1.02243328,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 2.3167057238875843,
      "language_loss": 0.7444104,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76649672,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.6124658584594727
    },
    {
      "auxiliary_loss_clip": 0.01180307,
      "auxiliary_loss_mlp": 0.01029248,
      "balance_loss_clip": 1.0568229,
      "balance_loss_mlp": 1.02121973,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.9341184554815678,
      "language_loss": 0.83200538,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85410094,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.5986363887786865
    },
    {
      "auxiliary_loss_clip": 0.01184412,
      "auxiliary_loss_mlp": 0.01031949,
      "balance_loss_clip": 1.05553961,
      "balance_loss_mlp": 1.02425444,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.0691206929429193,
      "language_loss": 0.70461184,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.72677541,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 3.5578417778015137
    },
    {
      "auxiliary_loss_clip": 0.01175943,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05228949,
      "balance_loss_mlp": 1.02221251,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 1.7007245123448513,
      "language_loss": 0.93990242,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.9619627,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.668668031692505
    },
    {
      "auxiliary_loss_clip": 0.0108639,
      "auxiliary_loss_mlp": 0.01003285,
      "balance_loss_clip": 1.02784681,
      "balance_loss_mlp": 1.0020808,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8634202908000597,
      "language_loss": 0.59814912,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61904585,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.034633159637451
    },
    {
      "auxiliary_loss_clip": 0.01173892,
      "auxiliary_loss_mlp": 0.01023049,
      "balance_loss_clip": 1.05411625,
      "balance_loss_mlp": 1.01445448,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 4.26513790383726,
      "language_loss": 0.7305572,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75252658,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.6268811225891113
    },
    {
      "auxiliary_loss_clip": 0.01188834,
      "auxiliary_loss_mlp": 0.01027438,
      "balance_loss_clip": 1.05734611,
      "balance_loss_mlp": 1.01929605,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.616563112445959,
      "language_loss": 0.68065894,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70282161,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.5282654762268066
    },
    {
      "auxiliary_loss_clip": 0.01149741,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 1.04586267,
      "balance_loss_mlp": 1.01879859,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.4826820327882944,
      "language_loss": 0.82809305,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84986269,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 3.5552148818969727
    },
    {
      "auxiliary_loss_clip": 0.01059596,
      "auxiliary_loss_mlp": 0.01004596,
      "balance_loss_clip": 1.02013421,
      "balance_loss_mlp": 1.0033741,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8353314584594009,
      "language_loss": 0.57383841,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59448034,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.957651138305664
    },
    {
      "auxiliary_loss_clip": 0.01172741,
      "auxiliary_loss_mlp": 0.01024066,
      "balance_loss_clip": 1.05293369,
      "balance_loss_mlp": 1.01621628,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.8929816585426915,
      "language_loss": 0.77216125,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79412937,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.6390607357025146
    },
    {
      "auxiliary_loss_clip": 0.0118312,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.05423307,
      "balance_loss_mlp": 1.01881933,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 3.399717757142657,
      "language_loss": 0.74297827,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76507676,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 3.505638599395752
    },
    {
      "auxiliary_loss_clip": 0.01179813,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.05441594,
      "balance_loss_mlp": 1.01618528,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.7254161695947696,
      "language_loss": 0.85352671,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87556589,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.610668659210205
    },
    {
      "auxiliary_loss_clip": 0.01141195,
      "auxiliary_loss_mlp": 0.01025519,
      "balance_loss_clip": 1.04737711,
      "balance_loss_mlp": 1.01720393,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 1.9387111901137328,
      "language_loss": 0.78591681,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80758393,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.76769757270813
    },
    {
      "auxiliary_loss_clip": 0.01173256,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.05200481,
      "balance_loss_mlp": 1.01663327,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.020814864231305,
      "language_loss": 0.78378057,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80576092,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.625
    },
    {
      "auxiliary_loss_clip": 0.01151194,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.04526758,
      "balance_loss_mlp": 1.01975346,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.1288665285293935,
      "language_loss": 0.80459237,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82637644,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 3.672903299331665
    },
    {
      "auxiliary_loss_clip": 0.01179513,
      "auxiliary_loss_mlp": 0.00901341,
      "balance_loss_clip": 1.05585599,
      "balance_loss_mlp": 1.00136936,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.232790926742613,
      "language_loss": 0.85466391,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87547243,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.562547445297241
    },
    {
      "auxiliary_loss_clip": 0.01191082,
      "auxiliary_loss_mlp": 0.01030207,
      "balance_loss_clip": 1.05606377,
      "balance_loss_mlp": 1.02167153,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 2.6952012740070947,
      "language_loss": 0.75908267,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.78129554,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.751749277114868
    },
    {
      "auxiliary_loss_clip": 0.01171661,
      "auxiliary_loss_mlp": 0.01034485,
      "balance_loss_clip": 1.0518719,
      "balance_loss_mlp": 1.02590227,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.7949977946735545,
      "language_loss": 0.87769294,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89975441,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.6368653774261475
    },
    {
      "auxiliary_loss_clip": 0.01183882,
      "auxiliary_loss_mlp": 0.01026203,
      "balance_loss_clip": 1.05470538,
      "balance_loss_mlp": 1.01806116,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 2.02495333252008,
      "language_loss": 0.69088745,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.71298832,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.6104817390441895
    },
    {
      "auxiliary_loss_clip": 0.01177487,
      "auxiliary_loss_mlp": 0.01033321,
      "balance_loss_clip": 1.05382848,
      "balance_loss_mlp": 1.02544141,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.737898988505615,
      "language_loss": 0.71660787,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.73871601,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.6776864528656006
    },
    {
      "auxiliary_loss_clip": 0.01132035,
      "auxiliary_loss_mlp": 0.0103625,
      "balance_loss_clip": 1.04739547,
      "balance_loss_mlp": 1.02737522,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 2.51726957884187,
      "language_loss": 0.70315695,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72483981,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.775045156478882
    },
    {
      "auxiliary_loss_clip": 0.01166697,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.05581403,
      "balance_loss_mlp": 1.01913297,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.6366465339995897,
      "language_loss": 0.82039833,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84234095,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.656379461288452
    },
    {
      "auxiliary_loss_clip": 0.0117387,
      "auxiliary_loss_mlp": 0.01031862,
      "balance_loss_clip": 1.05362391,
      "balance_loss_mlp": 1.02276599,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.071595374224287,
      "language_loss": 0.72202539,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74408275,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.6696505546569824
    },
    {
      "auxiliary_loss_clip": 0.0117462,
      "auxiliary_loss_mlp": 0.01033327,
      "balance_loss_clip": 1.05281186,
      "balance_loss_mlp": 1.02477419,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 5.7561281215953946,
      "language_loss": 0.84590501,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86798447,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.627309799194336
    },
    {
      "auxiliary_loss_clip": 0.01165562,
      "auxiliary_loss_mlp": 0.01023893,
      "balance_loss_clip": 1.04627824,
      "balance_loss_mlp": 1.01617694,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.923908947642861,
      "language_loss": 0.83414102,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85603559,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.6384310722351074
    },
    {
      "auxiliary_loss_clip": 0.01176674,
      "auxiliary_loss_mlp": 0.01028452,
      "balance_loss_clip": 1.05152011,
      "balance_loss_mlp": 1.02019024,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.7749567304093916,
      "language_loss": 0.74558669,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76763797,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.670156717300415
    },
    {
      "auxiliary_loss_clip": 0.01178207,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.05137801,
      "balance_loss_mlp": 1.02185726,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 3.4786381299047595,
      "language_loss": 0.79045618,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81254292,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.5806708335876465
    },
    {
      "auxiliary_loss_clip": 0.01186854,
      "auxiliary_loss_mlp": 0.01026134,
      "balance_loss_clip": 1.05367219,
      "balance_loss_mlp": 1.01731229,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 1.7287211139550516,
      "language_loss": 0.80770832,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82983828,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 2.641923427581787
    },
    {
      "auxiliary_loss_clip": 0.01176781,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.05508018,
      "balance_loss_mlp": 1.02043676,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 1.8673172015618864,
      "language_loss": 0.72376895,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.7458235,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.635561227798462
    },
    {
      "auxiliary_loss_clip": 0.01156743,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.04965925,
      "balance_loss_mlp": 1.02013922,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.6624642241406589,
      "language_loss": 0.7912128,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81306124,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 3.63379168510437
    },
    {
      "auxiliary_loss_clip": 0.01189084,
      "auxiliary_loss_mlp": 0.01026753,
      "balance_loss_clip": 1.05737138,
      "balance_loss_mlp": 1.01901317,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 2.4114590280601664,
      "language_loss": 0.78748047,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.8096388,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 2.565878391265869
    },
    {
      "auxiliary_loss_clip": 0.01178187,
      "auxiliary_loss_mlp": 0.01026807,
      "balance_loss_clip": 1.05367255,
      "balance_loss_mlp": 1.01846266,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 2.118162628663191,
      "language_loss": 0.88885844,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.9109084,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.5486137866973877
    },
    {
      "auxiliary_loss_clip": 0.01165838,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.0540843,
      "balance_loss_mlp": 1.01941001,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.9593420947730749,
      "language_loss": 0.86663485,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.8885662,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.682032585144043
    },
    {
      "auxiliary_loss_clip": 0.01155394,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.0451107,
      "balance_loss_mlp": 1.02262032,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.8808599685978524,
      "language_loss": 0.81430435,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83616829,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.8520450592041016
    },
    {
      "auxiliary_loss_clip": 0.01189412,
      "auxiliary_loss_mlp": 0.01030826,
      "balance_loss_clip": 1.05381358,
      "balance_loss_mlp": 1.02249277,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.594718493281332,
      "language_loss": 0.69772458,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71992695,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 3.541023015975952
    },
    {
      "auxiliary_loss_clip": 0.01173995,
      "auxiliary_loss_mlp": 0.01031871,
      "balance_loss_clip": 1.05179191,
      "balance_loss_mlp": 1.02390754,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.419535633050248,
      "language_loss": 0.81922072,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84127939,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.653881311416626
    },
    {
      "auxiliary_loss_clip": 0.01073435,
      "auxiliary_loss_mlp": 0.01000891,
      "balance_loss_clip": 1.01531696,
      "balance_loss_mlp": 0.99991304,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.9502202420919033,
      "language_loss": 0.56777155,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.5885148,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.292633533477783
    },
    {
      "auxiliary_loss_clip": 0.01182213,
      "auxiliary_loss_mlp": 0.00901465,
      "balance_loss_clip": 1.05461335,
      "balance_loss_mlp": 1.00134635,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.754233306105796,
      "language_loss": 0.71616578,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.73700255,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 3.7149178981781006
    },
    {
      "auxiliary_loss_clip": 0.0117811,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.05088484,
      "balance_loss_mlp": 1.01738787,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.9099011392204013,
      "language_loss": 0.73248935,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75452387,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.602910041809082
    },
    {
      "auxiliary_loss_clip": 0.01164274,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.04833913,
      "balance_loss_mlp": 1.01973104,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.7714871377185049,
      "language_loss": 0.9029634,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92488384,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.7330238819122314
    },
    {
      "auxiliary_loss_clip": 0.01165247,
      "auxiliary_loss_mlp": 0.01027853,
      "balance_loss_clip": 1.04982257,
      "balance_loss_mlp": 1.01997614,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.6215661815245788,
      "language_loss": 0.78925651,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81118751,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 3.6532294750213623
    },
    {
      "auxiliary_loss_clip": 0.01166769,
      "auxiliary_loss_mlp": 0.01031671,
      "balance_loss_clip": 1.05234194,
      "balance_loss_mlp": 1.02348137,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 2.270585939483348,
      "language_loss": 0.73647034,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.75845474,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.708890438079834
    },
    {
      "auxiliary_loss_clip": 0.01150604,
      "auxiliary_loss_mlp": 0.01033403,
      "balance_loss_clip": 1.04765379,
      "balance_loss_mlp": 1.02419972,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 1.950007257439563,
      "language_loss": 0.75320458,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77504468,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.692337989807129
    },
    {
      "auxiliary_loss_clip": 0.01177952,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.05524564,
      "balance_loss_mlp": 1.02171421,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.6387547939404128,
      "language_loss": 0.73306739,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.75514507,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.6135597229003906
    },
    {
      "auxiliary_loss_clip": 0.01188631,
      "auxiliary_loss_mlp": 0.01026524,
      "balance_loss_clip": 1.05478883,
      "balance_loss_mlp": 1.01840019,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 2.002555544477205,
      "language_loss": 0.79656148,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81871295,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.6456644535064697
    },
    {
      "auxiliary_loss_clip": 0.01085415,
      "auxiliary_loss_mlp": 0.0100103,
      "balance_loss_clip": 1.01634896,
      "balance_loss_mlp": 1.00003457,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.9198615725271871,
      "language_loss": 0.56883425,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.58969867,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.162860870361328
    },
    {
      "auxiliary_loss_clip": 0.01173052,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.0500083,
      "balance_loss_mlp": 1.01870799,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 3.0642141252699497,
      "language_loss": 0.6998291,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.72182381,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.7891223430633545
    },
    {
      "auxiliary_loss_clip": 0.01164012,
      "auxiliary_loss_mlp": 0.00901269,
      "balance_loss_clip": 1.04920268,
      "balance_loss_mlp": 1.00124645,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.351478286167268,
      "language_loss": 0.75800377,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.7786566,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.79050874710083
    },
    {
      "auxiliary_loss_clip": 0.01157453,
      "auxiliary_loss_mlp": 0.01032018,
      "balance_loss_clip": 1.04653478,
      "balance_loss_mlp": 1.02376223,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 2.108142176057433,
      "language_loss": 0.85326552,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87516022,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.6409783363342285
    },
    {
      "auxiliary_loss_clip": 0.01159398,
      "auxiliary_loss_mlp": 0.01036441,
      "balance_loss_clip": 1.0484885,
      "balance_loss_mlp": 1.02766168,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.2684587259725,
      "language_loss": 0.79000437,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.81196284,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.805469274520874
    },
    {
      "auxiliary_loss_clip": 0.01157033,
      "auxiliary_loss_mlp": 0.01024041,
      "balance_loss_clip": 1.04834378,
      "balance_loss_mlp": 1.01613736,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 2.0672246819589635,
      "language_loss": 0.79348159,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.8152923,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.818049669265747
    },
    {
      "auxiliary_loss_clip": 0.01188969,
      "auxiliary_loss_mlp": 0.0102429,
      "balance_loss_clip": 1.05519342,
      "balance_loss_mlp": 1.01650524,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 3.280048292236222,
      "language_loss": 0.84271562,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86484814,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.611361265182495
    },
    {
      "auxiliary_loss_clip": 0.01191399,
      "auxiliary_loss_mlp": 0.01031717,
      "balance_loss_clip": 1.0538274,
      "balance_loss_mlp": 1.02284241,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 2.773521575752653,
      "language_loss": 0.84857655,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.87080765,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.635833740234375
    },
    {
      "auxiliary_loss_clip": 0.01170819,
      "auxiliary_loss_mlp": 0.0090218,
      "balance_loss_clip": 1.05289984,
      "balance_loss_mlp": 1.00141037,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.8461732795393933,
      "language_loss": 0.81779432,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83852434,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 2.662541151046753
    },
    {
      "auxiliary_loss_clip": 0.01073194,
      "auxiliary_loss_mlp": 0.01002032,
      "balance_loss_clip": 1.01588178,
      "balance_loss_mlp": 1.00103617,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8231756603873517,
      "language_loss": 0.60949343,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.63024569,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.2663300037384033
    },
    {
      "auxiliary_loss_clip": 0.0113909,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.0458926,
      "balance_loss_mlp": 1.02039886,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.0465412530487397,
      "language_loss": 0.78706533,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.80874258,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 3.6921048164367676
    },
    {
      "auxiliary_loss_clip": 0.01156621,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.04906857,
      "balance_loss_mlp": 1.01841903,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 2.326820818067988,
      "language_loss": 0.84803838,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.86986673,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.666476249694824
    },
    {
      "auxiliary_loss_clip": 0.01176694,
      "auxiliary_loss_mlp": 0.01029099,
      "balance_loss_clip": 1.05102956,
      "balance_loss_mlp": 1.02131486,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.91176426957735,
      "language_loss": 0.84671021,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86876816,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.646094799041748
    },
    {
      "auxiliary_loss_clip": 0.01176498,
      "auxiliary_loss_mlp": 0.0103383,
      "balance_loss_clip": 1.05644238,
      "balance_loss_mlp": 1.02497911,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 3.3518221665838737,
      "language_loss": 0.84164095,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86374426,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.7744569778442383
    },
    {
      "auxiliary_loss_clip": 0.01190428,
      "auxiliary_loss_mlp": 0.01027306,
      "balance_loss_clip": 1.05702901,
      "balance_loss_mlp": 1.01903284,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.4762146609971103,
      "language_loss": 0.78553224,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80770957,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.5851287841796875
    },
    {
      "auxiliary_loss_clip": 0.01179227,
      "auxiliary_loss_mlp": 0.01030439,
      "balance_loss_clip": 1.0530262,
      "balance_loss_mlp": 1.02121806,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 1.8551716930724367,
      "language_loss": 0.74056756,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.7626642,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.6552984714508057
    },
    {
      "auxiliary_loss_clip": 0.01141979,
      "auxiliary_loss_mlp": 0.01031917,
      "balance_loss_clip": 1.04485643,
      "balance_loss_mlp": 1.02368605,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 1.7277160771804372,
      "language_loss": 0.82031101,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.84204996,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 3.6689889430999756
    },
    {
      "auxiliary_loss_clip": 0.01179605,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.0557332,
      "balance_loss_mlp": 1.01950228,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.410537913479052,
      "language_loss": 0.81921518,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.84128606,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.673027515411377
    },
    {
      "auxiliary_loss_clip": 0.01177752,
      "auxiliary_loss_mlp": 0.00901799,
      "balance_loss_clip": 1.05373442,
      "balance_loss_mlp": 1.00128841,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 2.204896331044578,
      "language_loss": 0.71228999,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73308551,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 3.714350700378418
    },
    {
      "auxiliary_loss_clip": 0.01182866,
      "auxiliary_loss_mlp": 0.01033683,
      "balance_loss_clip": 1.05149639,
      "balance_loss_mlp": 1.0242362,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 2.841771002311313,
      "language_loss": 0.83170867,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85387421,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.7451534271240234
    },
    {
      "auxiliary_loss_clip": 0.01174394,
      "auxiliary_loss_mlp": 0.01035901,
      "balance_loss_clip": 1.05463505,
      "balance_loss_mlp": 1.02790236,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.6533484836116707,
      "language_loss": 0.83674097,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.85884386,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.6490187644958496
    },
    {
      "auxiliary_loss_clip": 0.01177485,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.05283546,
      "balance_loss_mlp": 1.02189732,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 1.986767493369407,
      "language_loss": 0.79701066,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.81908655,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.7473485469818115
    },
    {
      "auxiliary_loss_clip": 0.01191891,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.0586797,
      "balance_loss_mlp": 1.01915824,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.7710263891843379,
      "language_loss": 0.79876804,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82095718,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 3.5015196800231934
    },
    {
      "auxiliary_loss_clip": 0.01170556,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05419326,
      "balance_loss_mlp": 1.02210391,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 1.8564256518106168,
      "language_loss": 0.7272476,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.74924958,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.692913293838501
    },
    {
      "auxiliary_loss_clip": 0.01174529,
      "auxiliary_loss_mlp": 0.01028923,
      "balance_loss_clip": 1.05377543,
      "balance_loss_mlp": 1.02072692,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 2.1637393676772247,
      "language_loss": 0.78527999,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80731452,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.6007182598114014
    },
    {
      "auxiliary_loss_clip": 0.01160605,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.04791427,
      "balance_loss_mlp": 1.02106094,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.4091217301110124,
      "language_loss": 0.86876273,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89066076,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.711529016494751
    },
    {
      "auxiliary_loss_clip": 0.01180517,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.05374694,
      "balance_loss_mlp": 1.0186429,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.1166818438132315,
      "language_loss": 0.74475265,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.76682007,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.54146146774292
    },
    {
      "auxiliary_loss_clip": 0.01178627,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.05460846,
      "balance_loss_mlp": 1.01589835,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 2.695925288133114,
      "language_loss": 0.82505077,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84708184,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6324663162231445
    },
    {
      "auxiliary_loss_clip": 0.0115935,
      "auxiliary_loss_mlp": 0.01026893,
      "balance_loss_clip": 1.05328655,
      "balance_loss_mlp": 1.01860237,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.6971188955208867,
      "language_loss": 0.67714405,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.6990065,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.688079595565796
    },
    {
      "auxiliary_loss_clip": 0.0118795,
      "auxiliary_loss_mlp": 0.01029536,
      "balance_loss_clip": 1.05372691,
      "balance_loss_mlp": 1.02129889,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.9632909146612958,
      "language_loss": 0.75629401,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77846891,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.680955171585083
    },
    {
      "auxiliary_loss_clip": 0.01170521,
      "auxiliary_loss_mlp": 0.00901416,
      "balance_loss_clip": 1.0548017,
      "balance_loss_mlp": 1.00113738,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.0116148001709124,
      "language_loss": 0.71091312,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73163247,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.76631760597229
    },
    {
      "auxiliary_loss_clip": 0.01160072,
      "auxiliary_loss_mlp": 0.0102419,
      "balance_loss_clip": 1.04578447,
      "balance_loss_mlp": 1.01674557,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 2.228210844302113,
      "language_loss": 0.7056964,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.727539,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.6768717765808105
    },
    {
      "auxiliary_loss_clip": 0.01144822,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04744816,
      "balance_loss_mlp": 1.01900721,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 2.287285781104856,
      "language_loss": 0.70521694,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72693038,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.9829463958740234
    },
    {
      "auxiliary_loss_clip": 0.01162897,
      "auxiliary_loss_mlp": 0.01025807,
      "balance_loss_clip": 1.05322862,
      "balance_loss_mlp": 1.01806402,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 1.8568974381095136,
      "language_loss": 0.70140564,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.72329265,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.7599544525146484
    },
    {
      "auxiliary_loss_clip": 0.01169166,
      "auxiliary_loss_mlp": 0.01033839,
      "balance_loss_clip": 1.05178404,
      "balance_loss_mlp": 1.02548265,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.6366363859155708,
      "language_loss": 0.85198146,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87401152,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.7253336906433105
    },
    {
      "auxiliary_loss_clip": 0.0118935,
      "auxiliary_loss_mlp": 0.01025999,
      "balance_loss_clip": 1.05895138,
      "balance_loss_mlp": 1.01734495,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 4.087809238510967,
      "language_loss": 0.74634004,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76849353,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.5637409687042236
    },
    {
      "auxiliary_loss_clip": 0.01189342,
      "auxiliary_loss_mlp": 0.0102573,
      "balance_loss_clip": 1.0556426,
      "balance_loss_mlp": 1.01692665,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 2.879324067489405,
      "language_loss": 0.69639409,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71854484,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 3.5051043033599854
    },
    {
      "auxiliary_loss_clip": 0.01180681,
      "auxiliary_loss_mlp": 0.01028545,
      "balance_loss_clip": 1.05513334,
      "balance_loss_mlp": 1.02068663,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.2125841985296413,
      "language_loss": 0.68837309,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.71046543,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.5855212211608887
    },
    {
      "auxiliary_loss_clip": 0.0118795,
      "auxiliary_loss_mlp": 0.01026644,
      "balance_loss_clip": 1.05428028,
      "balance_loss_mlp": 1.01859725,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 3.747782456360445,
      "language_loss": 0.74010313,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.76224905,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.5821194648742676
    },
    {
      "auxiliary_loss_clip": 0.01188235,
      "auxiliary_loss_mlp": 0.01024574,
      "balance_loss_clip": 1.05616057,
      "balance_loss_mlp": 1.01674771,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.479863566736964,
      "language_loss": 0.67409456,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.6962226,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.5274641513824463
    },
    {
      "auxiliary_loss_clip": 0.01154895,
      "auxiliary_loss_mlp": 0.01033973,
      "balance_loss_clip": 1.04935193,
      "balance_loss_mlp": 1.02503812,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 2.272720991000587,
      "language_loss": 0.8203336,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84222233,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.7025346755981445
    },
    {
      "auxiliary_loss_clip": 0.01185777,
      "auxiliary_loss_mlp": 0.00901387,
      "balance_loss_clip": 1.0565635,
      "balance_loss_mlp": 1.00104415,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.8780557368650277,
      "language_loss": 0.70118487,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72205651,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.6301751136779785
    },
    {
      "auxiliary_loss_clip": 0.01194427,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.05869007,
      "balance_loss_mlp": 1.02155495,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.8266717764518945,
      "language_loss": 0.62908453,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65132618,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 3.58931040763855
    },
    {
      "auxiliary_loss_clip": 0.01165123,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.05219007,
      "balance_loss_mlp": 1.02354932,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.7186281373205936,
      "language_loss": 0.65651703,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67848206,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.6895432472229004
    },
    {
      "auxiliary_loss_clip": 0.01142873,
      "auxiliary_loss_mlp": 0.0102538,
      "balance_loss_clip": 1.0474596,
      "balance_loss_mlp": 1.01828718,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.6779376534617618,
      "language_loss": 0.74549913,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76718163,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.6848273277282715
    },
    {
      "auxiliary_loss_clip": 0.01170082,
      "auxiliary_loss_mlp": 0.01030463,
      "balance_loss_clip": 1.05431032,
      "balance_loss_mlp": 1.02239799,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.7812451031260315,
      "language_loss": 0.80912089,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.83112633,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 3.6306307315826416
    },
    {
      "auxiliary_loss_clip": 0.0116363,
      "auxiliary_loss_mlp": 0.01023991,
      "balance_loss_clip": 1.05394363,
      "balance_loss_mlp": 1.01607549,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.1303899601880003,
      "language_loss": 0.68556309,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.70743936,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.598423957824707
    },
    {
      "auxiliary_loss_clip": 0.01162561,
      "auxiliary_loss_mlp": 0.00901571,
      "balance_loss_clip": 1.05143166,
      "balance_loss_mlp": 1.00100613,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.436223147983307,
      "language_loss": 0.80549341,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82613468,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.6681010723114014
    },
    {
      "auxiliary_loss_clip": 0.01175689,
      "auxiliary_loss_mlp": 0.0102614,
      "balance_loss_clip": 1.05165076,
      "balance_loss_mlp": 1.01892138,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.8300917751450927,
      "language_loss": 0.78329432,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80531257,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.6282541751861572
    },
    {
      "auxiliary_loss_clip": 0.01182505,
      "auxiliary_loss_mlp": 0.01033048,
      "balance_loss_clip": 1.0557847,
      "balance_loss_mlp": 1.02462602,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.6004544628446027,
      "language_loss": 0.80140519,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.82356071,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 3.513654947280884
    },
    {
      "auxiliary_loss_clip": 0.01188171,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.05704641,
      "balance_loss_mlp": 1.01758885,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 3.4354506662127595,
      "language_loss": 0.86359459,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88573098,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.6653451919555664
    },
    {
      "auxiliary_loss_clip": 0.01163419,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.05490458,
      "balance_loss_mlp": 1.01692379,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 2.281309724439514,
      "language_loss": 0.75315142,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77503568,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.6812009811401367
    },
    {
      "auxiliary_loss_clip": 0.01186935,
      "auxiliary_loss_mlp": 0.01024771,
      "balance_loss_clip": 1.05450916,
      "balance_loss_mlp": 1.01632237,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.8021784108367322,
      "language_loss": 0.71568197,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73779905,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.681609630584717
    },
    {
      "auxiliary_loss_clip": 0.01155018,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 1.05152667,
      "balance_loss_mlp": 1.01874781,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 1.7839799301305066,
      "language_loss": 0.57320642,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.59501946,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.7714109420776367
    },
    {
      "auxiliary_loss_clip": 0.01165687,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.0506295,
      "balance_loss_mlp": 1.01928282,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 2.649393744227641,
      "language_loss": 0.69183362,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71376204,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.656012535095215
    },
    {
      "auxiliary_loss_clip": 0.01190684,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.05777788,
      "balance_loss_mlp": 1.02264833,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.9930661312183797,
      "language_loss": 0.73106217,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75328088,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.716848373413086
    },
    {
      "auxiliary_loss_clip": 0.0117337,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.05340362,
      "balance_loss_mlp": 1.01784873,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.8647740123835432,
      "language_loss": 0.7813471,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.80333704,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.6721584796905518
    },
    {
      "auxiliary_loss_clip": 0.01165446,
      "auxiliary_loss_mlp": 0.01029024,
      "balance_loss_clip": 1.05256486,
      "balance_loss_mlp": 1.02104902,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.824906519933992,
      "language_loss": 0.74070501,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.76264977,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.742896795272827
    },
    {
      "auxiliary_loss_clip": 0.01162342,
      "auxiliary_loss_mlp": 0.01024556,
      "balance_loss_clip": 1.05126345,
      "balance_loss_mlp": 1.01675367,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 2.039122997761628,
      "language_loss": 0.70865536,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.7305243,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.725294351577759
    },
    {
      "auxiliary_loss_clip": 0.01179228,
      "auxiliary_loss_mlp": 0.01030655,
      "balance_loss_clip": 1.05540705,
      "balance_loss_mlp": 1.02256107,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.2374491129009058,
      "language_loss": 0.83048892,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.85258782,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.648411750793457
    },
    {
      "auxiliary_loss_clip": 0.01161607,
      "auxiliary_loss_mlp": 0.01023364,
      "balance_loss_clip": 1.05171442,
      "balance_loss_mlp": 1.01531756,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 3.2634512295704625,
      "language_loss": 0.76333237,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.78518206,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 2.8798022270202637
    },
    {
      "auxiliary_loss_clip": 0.01137041,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.04510486,
      "balance_loss_mlp": 1.01666951,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 1.9662455079707697,
      "language_loss": 0.79268241,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81429654,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.725529432296753
    },
    {
      "auxiliary_loss_clip": 0.01182722,
      "auxiliary_loss_mlp": 0.01033369,
      "balance_loss_clip": 1.05600441,
      "balance_loss_mlp": 1.02476835,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.8818999461350716,
      "language_loss": 0.68829376,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.7104547,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 2.847994327545166
    },
    {
      "auxiliary_loss_clip": 0.01169498,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.05542386,
      "balance_loss_mlp": 1.02106094,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.5314313197545493,
      "language_loss": 0.71952927,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.74151522,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 3.7214996814727783
    },
    {
      "auxiliary_loss_clip": 0.01182288,
      "auxiliary_loss_mlp": 0.01032874,
      "balance_loss_clip": 1.05674219,
      "balance_loss_mlp": 1.02459526,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.7234561543557474,
      "language_loss": 0.77367997,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79583162,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.7325375080108643
    },
    {
      "auxiliary_loss_clip": 0.01188065,
      "auxiliary_loss_mlp": 0.00901375,
      "balance_loss_clip": 1.0544225,
      "balance_loss_mlp": 1.00094521,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.6631659720571577,
      "language_loss": 0.888708,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.9096024,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 2.5328705310821533
    },
    {
      "auxiliary_loss_clip": 0.01174305,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.05182159,
      "balance_loss_mlp": 1.01824605,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.7305062384688321,
      "language_loss": 0.81940913,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.84141648,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.6991686820983887
    },
    {
      "auxiliary_loss_clip": 0.01187773,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.05598474,
      "balance_loss_mlp": 1.01810527,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 2.0057873682673324,
      "language_loss": 0.72961539,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.7517575,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.6186985969543457
    },
    {
      "auxiliary_loss_clip": 0.01162944,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.0539819,
      "balance_loss_mlp": 1.02031827,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.8212339394600012,
      "language_loss": 0.84841985,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87033761,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 3.685927152633667
    },
    {
      "auxiliary_loss_clip": 0.01180621,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.0583657,
      "balance_loss_mlp": 1.02238142,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 2.055887204126565,
      "language_loss": 0.84254193,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86465347,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.6268537044525146
    },
    {
      "auxiliary_loss_clip": 0.01191116,
      "auxiliary_loss_mlp": 0.01031892,
      "balance_loss_clip": 1.05965745,
      "balance_loss_mlp": 1.02381563,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 3.702537400813935,
      "language_loss": 0.78780371,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.81003374,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.540544033050537
    },
    {
      "auxiliary_loss_clip": 0.01178446,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.05429459,
      "balance_loss_mlp": 1.01704717,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.7229636282078742,
      "language_loss": 0.65861303,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.68065208,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 3.6677603721618652
    },
    {
      "auxiliary_loss_clip": 0.01149588,
      "auxiliary_loss_mlp": 0.01025402,
      "balance_loss_clip": 1.05000043,
      "balance_loss_mlp": 1.01751029,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.397155799864524,
      "language_loss": 0.75038666,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77213657,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.7007031440734863
    },
    {
      "auxiliary_loss_clip": 0.01183146,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.05516791,
      "balance_loss_mlp": 1.0205946,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.0825491152384608,
      "language_loss": 0.72811627,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.75022995,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.644073247909546
    },
    {
      "auxiliary_loss_clip": 0.01155748,
      "auxiliary_loss_mlp": 0.01032735,
      "balance_loss_clip": 1.05247426,
      "balance_loss_mlp": 1.02488542,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 3.0154868202171157,
      "language_loss": 0.73085892,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75274372,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.643935441970825
    },
    {
      "auxiliary_loss_clip": 0.0114903,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.05175436,
      "balance_loss_mlp": 1.02153134,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 12.57759721561368,
      "language_loss": 0.81494713,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83673954,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.6980631351470947
    },
    {
      "auxiliary_loss_clip": 0.01144674,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.0488646,
      "balance_loss_mlp": 1.01975727,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 1.9296740284637166,
      "language_loss": 0.83466351,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85638654,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 3.6815218925476074
    },
    {
      "auxiliary_loss_clip": 0.01078621,
      "auxiliary_loss_mlp": 0.01004902,
      "balance_loss_clip": 1.02157116,
      "balance_loss_mlp": 1.00399005,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.7401625524843954,
      "language_loss": 0.58163667,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60247195,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.343939781188965
    },
    {
      "auxiliary_loss_clip": 0.01065485,
      "auxiliary_loss_mlp": 0.01001964,
      "balance_loss_clip": 1.02423668,
      "balance_loss_mlp": 1.00080764,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7848218705030071,
      "language_loss": 0.59842122,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61909568,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.351832628250122
    },
    {
      "auxiliary_loss_clip": 0.01145515,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.04732609,
      "balance_loss_mlp": 1.02174282,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 2.6585160478993197,
      "language_loss": 0.80226445,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82401097,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.7065885066986084
    },
    {
      "auxiliary_loss_clip": 0.0117302,
      "auxiliary_loss_mlp": 0.01034901,
      "balance_loss_clip": 1.05221021,
      "balance_loss_mlp": 1.0263958,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 5.428797948531016,
      "language_loss": 0.74903953,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77111882,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.60730242729187
    },
    {
      "auxiliary_loss_clip": 0.01171539,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.05269623,
      "balance_loss_mlp": 1.01729488,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.050883242887791,
      "language_loss": 0.72035152,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74232912,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.685774564743042
    },
    {
      "auxiliary_loss_clip": 0.01187848,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.05794477,
      "balance_loss_mlp": 1.01737833,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 2.805240972022512,
      "language_loss": 0.86762625,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.88975811,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.5525481700897217
    },
    {
      "auxiliary_loss_clip": 0.01173421,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.05685139,
      "balance_loss_mlp": 1.0248158,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.5423642841362974,
      "language_loss": 0.76548111,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78754139,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.7385382652282715
    },
    {
      "auxiliary_loss_clip": 0.0116041,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.05349851,
      "balance_loss_mlp": 1.02325225,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.6591605939602037,
      "language_loss": 0.84189808,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86382163,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.7584195137023926
    },
    {
      "auxiliary_loss_clip": 0.01176862,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.05509102,
      "balance_loss_mlp": 1.01558185,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 2.4771562874461583,
      "language_loss": 0.82914919,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.85116106,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.6352956295013428
    },
    {
      "auxiliary_loss_clip": 0.01191389,
      "auxiliary_loss_mlp": 0.00901769,
      "balance_loss_clip": 1.05750644,
      "balance_loss_mlp": 1.00092769,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.694869410228743,
      "language_loss": 0.75879234,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.77972388,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.5341837406158447
    },
    {
      "auxiliary_loss_clip": 0.01178804,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.05662513,
      "balance_loss_mlp": 1.02071929,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 3.7940637596875972,
      "language_loss": 0.8394655,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86154878,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.727675199508667
    },
    {
      "auxiliary_loss_clip": 0.01162934,
      "auxiliary_loss_mlp": 0.01023213,
      "balance_loss_clip": 1.05184054,
      "balance_loss_mlp": 1.01531267,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 2.0701834844726297,
      "language_loss": 0.7717576,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79361916,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.6382060050964355
    },
    {
      "auxiliary_loss_clip": 0.01190748,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 1.05663955,
      "balance_loss_mlp": 1.01998234,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.6626188678716471,
      "language_loss": 0.78365946,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.8058446,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.695441484451294
    },
    {
      "auxiliary_loss_clip": 0.01181428,
      "auxiliary_loss_mlp": 0.0102475,
      "balance_loss_clip": 1.05173361,
      "balance_loss_mlp": 1.01667356,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.9581661565815205,
      "language_loss": 0.78649265,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80855441,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 3.6093082427978516
    },
    {
      "auxiliary_loss_clip": 0.01168674,
      "auxiliary_loss_mlp": 0.01031592,
      "balance_loss_clip": 1.05134797,
      "balance_loss_mlp": 1.02303922,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 2.4721974048396236,
      "language_loss": 0.73735118,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.75935382,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 2.6235668659210205
    },
    {
      "auxiliary_loss_clip": 0.0118487,
      "auxiliary_loss_mlp": 0.0103,
      "balance_loss_clip": 1.05566347,
      "balance_loss_mlp": 1.02232051,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.6182615703220975,
      "language_loss": 0.87042952,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89257824,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.6915621757507324
    },
    {
      "auxiliary_loss_clip": 0.0117828,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.05711293,
      "balance_loss_mlp": 1.01806688,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 2.0433880243143405,
      "language_loss": 0.81491554,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83695602,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.6726558208465576
    },
    {
      "auxiliary_loss_clip": 0.01179247,
      "auxiliary_loss_mlp": 0.01027753,
      "balance_loss_clip": 1.0563941,
      "balance_loss_mlp": 1.02043986,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.8818536072892247,
      "language_loss": 0.76150823,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78357816,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.6352598667144775
    },
    {
      "auxiliary_loss_clip": 0.01163822,
      "auxiliary_loss_mlp": 0.01033279,
      "balance_loss_clip": 1.05117202,
      "balance_loss_mlp": 1.02466643,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 4.414654573068284,
      "language_loss": 0.78850603,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81047702,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 3.505951404571533
    },
    {
      "auxiliary_loss_clip": 0.01166094,
      "auxiliary_loss_mlp": 0.01026399,
      "balance_loss_clip": 1.0527246,
      "balance_loss_mlp": 1.01841223,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.798653897547026,
      "language_loss": 0.83712065,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85904557,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.7426984310150146
    },
    {
      "auxiliary_loss_clip": 0.01193838,
      "auxiliary_loss_mlp": 0.01023468,
      "balance_loss_clip": 1.06068993,
      "balance_loss_mlp": 1.01542151,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 2.3339071665918363,
      "language_loss": 0.68103445,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70320749,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.6045639514923096
    },
    {
      "auxiliary_loss_clip": 0.01191076,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.05885458,
      "balance_loss_mlp": 1.01994956,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.256131516074863,
      "language_loss": 0.72773308,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.74992681,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 3.520787477493286
    },
    {
      "auxiliary_loss_clip": 0.0116954,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.05213428,
      "balance_loss_mlp": 1.01817369,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.6060222601733973,
      "language_loss": 0.68203777,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70400244,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.887869119644165
    },
    {
      "auxiliary_loss_clip": 0.01155148,
      "auxiliary_loss_mlp": 0.01036228,
      "balance_loss_clip": 1.05388689,
      "balance_loss_mlp": 1.02767491,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 3.1544531843459303,
      "language_loss": 0.76784134,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.78975511,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.6960036754608154
    },
    {
      "auxiliary_loss_clip": 0.01171715,
      "auxiliary_loss_mlp": 0.01034436,
      "balance_loss_clip": 1.05439854,
      "balance_loss_mlp": 1.02579284,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.1287223696571407,
      "language_loss": 0.85033923,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87240076,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.6348636150360107
    },
    {
      "auxiliary_loss_clip": 0.01162147,
      "auxiliary_loss_mlp": 0.01025147,
      "balance_loss_clip": 1.05570912,
      "balance_loss_mlp": 1.01629543,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.211349567029898,
      "language_loss": 0.81146145,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83333433,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.6392550468444824
    },
    {
      "auxiliary_loss_clip": 0.01164128,
      "auxiliary_loss_mlp": 0.0103273,
      "balance_loss_clip": 1.05348694,
      "balance_loss_mlp": 1.02377152,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 2.390986695704807,
      "language_loss": 0.70562971,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72759831,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 3.703146457672119
    },
    {
      "auxiliary_loss_clip": 0.01161173,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.05508244,
      "balance_loss_mlp": 1.0193069,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.746697903032036,
      "language_loss": 0.83360088,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85548329,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.7063164710998535
    },
    {
      "auxiliary_loss_clip": 0.01155683,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.05496883,
      "balance_loss_mlp": 1.02352417,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 2.166217970696034,
      "language_loss": 0.82302845,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84490269,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.7170047760009766
    },
    {
      "auxiliary_loss_clip": 0.01144736,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.04857218,
      "balance_loss_mlp": 1.02075171,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 2.9777217495518333,
      "language_loss": 0.83451313,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85624874,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.759201765060425
    },
    {
      "auxiliary_loss_clip": 0.01147301,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.04620361,
      "balance_loss_mlp": 1.02195835,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.7833463286620967,
      "language_loss": 0.77598619,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79775542,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.771305799484253
    },
    {
      "auxiliary_loss_clip": 0.01180652,
      "auxiliary_loss_mlp": 0.01028696,
      "balance_loss_clip": 1.05300474,
      "balance_loss_mlp": 1.02022636,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.6698385826647995,
      "language_loss": 0.78397679,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80607021,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.6053104400634766
    },
    {
      "auxiliary_loss_clip": 0.01098469,
      "auxiliary_loss_mlp": 0.01001773,
      "balance_loss_clip": 1.02825546,
      "balance_loss_mlp": 1.00071228,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7959536268446755,
      "language_loss": 0.62435782,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64536023,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.2384681701660156
    },
    {
      "auxiliary_loss_clip": 0.01171453,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.0573988,
      "balance_loss_mlp": 1.02020693,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.414612031763583,
      "language_loss": 0.85350376,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.8755073,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.7256710529327393
    },
    {
      "auxiliary_loss_clip": 0.01167172,
      "auxiliary_loss_mlp": 0.01023087,
      "balance_loss_clip": 1.05318248,
      "balance_loss_mlp": 1.01486146,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 2.495604261616195,
      "language_loss": 0.82013214,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.8420347,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.722612142562866
    },
    {
      "auxiliary_loss_clip": 0.01164586,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.05568457,
      "balance_loss_mlp": 1.01786482,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 1.9321854582622602,
      "language_loss": 0.74690777,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76882404,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.7269959449768066
    },
    {
      "auxiliary_loss_clip": 0.0109352,
      "auxiliary_loss_mlp": 0.01001826,
      "balance_loss_clip": 1.02716398,
      "balance_loss_mlp": 1.00077677,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7939199699126436,
      "language_loss": 0.62916052,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.65011394,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.2696735858917236
    },
    {
      "auxiliary_loss_clip": 0.01173726,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.05762458,
      "balance_loss_mlp": 1.01700878,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.27249875559496,
      "language_loss": 0.72790259,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74989396,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.6697826385498047
    },
    {
      "auxiliary_loss_clip": 0.01191693,
      "auxiliary_loss_mlp": 0.01023211,
      "balance_loss_clip": 1.06001496,
      "balance_loss_mlp": 1.01551914,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 1.836996651702883,
      "language_loss": 0.81006908,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83221811,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.703603744506836
    },
    {
      "auxiliary_loss_clip": 0.01185735,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.05675519,
      "balance_loss_mlp": 1.02002656,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 2.0999403998803787,
      "language_loss": 0.734761,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75690663,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 3.5700056552886963
    },
    {
      "auxiliary_loss_clip": 0.01177918,
      "auxiliary_loss_mlp": 0.01030752,
      "balance_loss_clip": 1.05311871,
      "balance_loss_mlp": 1.02234733,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 1.8969794407511964,
      "language_loss": 0.7607075,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78279424,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.644481658935547
    },
    {
      "auxiliary_loss_clip": 0.01180582,
      "auxiliary_loss_mlp": 0.01033292,
      "balance_loss_clip": 1.05535007,
      "balance_loss_mlp": 1.02565384,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 2.0833132391397733,
      "language_loss": 0.90408719,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.92622602,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.7091140747070312
    },
    {
      "auxiliary_loss_clip": 0.01181907,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.05722082,
      "balance_loss_mlp": 1.0198257,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.6673975182398804,
      "language_loss": 0.80493957,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82703948,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.633946418762207
    },
    {
      "auxiliary_loss_clip": 0.01177667,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.05286241,
      "balance_loss_mlp": 1.02035999,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.0810047982752877,
      "language_loss": 0.74242812,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76448911,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.69132399559021
    },
    {
      "auxiliary_loss_clip": 0.01189719,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.05780911,
      "balance_loss_mlp": 1.01799011,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 2.041207780309398,
      "language_loss": 0.76741827,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78958142,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.692894458770752
    },
    {
      "auxiliary_loss_clip": 0.01179206,
      "auxiliary_loss_mlp": 0.00902262,
      "balance_loss_clip": 1.05843759,
      "balance_loss_mlp": 1.00101852,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 2.1840373664891013,
      "language_loss": 0.82288814,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.84370279,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 3.61273193359375
    },
    {
      "auxiliary_loss_clip": 0.0117709,
      "auxiliary_loss_mlp": 0.01031602,
      "balance_loss_clip": 1.05050755,
      "balance_loss_mlp": 1.02329326,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.590723121709932,
      "language_loss": 0.66320133,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68528831,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.5770621299743652
    },
    {
      "auxiliary_loss_clip": 0.0118318,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.05238008,
      "balance_loss_mlp": 1.01561713,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 1.6513270866213206,
      "language_loss": 0.79321003,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81528401,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.701509475708008
    },
    {
      "auxiliary_loss_clip": 0.01165339,
      "auxiliary_loss_mlp": 0.01028895,
      "balance_loss_clip": 1.05255628,
      "balance_loss_mlp": 1.02044952,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.817079766301132,
      "language_loss": 0.86009252,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.88203478,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 3.596160411834717
    },
    {
      "auxiliary_loss_clip": 0.01156525,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.04772425,
      "balance_loss_mlp": 1.01785767,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 4.281417894593526,
      "language_loss": 0.68141794,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.70324856,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.6608245372772217
    },
    {
      "auxiliary_loss_clip": 0.01166127,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.05360174,
      "balance_loss_mlp": 1.02683115,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.7444766448989506,
      "language_loss": 0.82625544,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84826875,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.6852500438690186
    },
    {
      "auxiliary_loss_clip": 0.01180445,
      "auxiliary_loss_mlp": 0.00901592,
      "balance_loss_clip": 1.05739975,
      "balance_loss_mlp": 1.00102258,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.5789332707365844,
      "language_loss": 0.67061299,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.69143337,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.681499719619751
    },
    {
      "auxiliary_loss_clip": 0.01159461,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.05111015,
      "balance_loss_mlp": 1.01668119,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 1.8397911899487858,
      "language_loss": 0.79511464,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81696576,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 3.6392483711242676
    },
    {
      "auxiliary_loss_clip": 0.01169024,
      "auxiliary_loss_mlp": 0.01030702,
      "balance_loss_clip": 1.05447459,
      "balance_loss_mlp": 1.02227998,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.6638552047089992,
      "language_loss": 0.7276271,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.74962437,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.6264429092407227
    },
    {
      "auxiliary_loss_clip": 0.01179813,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 1.05337,
      "balance_loss_mlp": 1.025666,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.8826268895427904,
      "language_loss": 0.60823369,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63037729,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.7730588912963867
    },
    {
      "auxiliary_loss_clip": 0.01178575,
      "auxiliary_loss_mlp": 0.01031892,
      "balance_loss_clip": 1.05537736,
      "balance_loss_mlp": 1.02298069,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 2.078813788909686,
      "language_loss": 0.79623455,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81833923,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.619990110397339
    },
    {
      "auxiliary_loss_clip": 0.01178804,
      "auxiliary_loss_mlp": 0.0103085,
      "balance_loss_clip": 1.05747759,
      "balance_loss_mlp": 1.02299976,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.5717745261418101,
      "language_loss": 0.81618714,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83828372,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.6271350383758545
    },
    {
      "auxiliary_loss_clip": 0.01181197,
      "auxiliary_loss_mlp": 0.01031868,
      "balance_loss_clip": 1.05730605,
      "balance_loss_mlp": 1.02283835,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 2.0019333546593896,
      "language_loss": 0.72869253,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.7508232,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.677978038787842
    },
    {
      "auxiliary_loss_clip": 0.0118526,
      "auxiliary_loss_mlp": 0.00902356,
      "balance_loss_clip": 1.05759215,
      "balance_loss_mlp": 1.00111985,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 2.7036127913136445,
      "language_loss": 0.91290474,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93378091,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.6733264923095703
    },
    {
      "auxiliary_loss_clip": 0.01175372,
      "auxiliary_loss_mlp": 0.01031739,
      "balance_loss_clip": 1.05352736,
      "balance_loss_mlp": 1.02354336,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 2.049477358473483,
      "language_loss": 0.75878584,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78085697,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.6282739639282227
    },
    {
      "auxiliary_loss_clip": 0.01180989,
      "auxiliary_loss_mlp": 0.01030785,
      "balance_loss_clip": 1.05643225,
      "balance_loss_mlp": 1.02263069,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.5141733433217415,
      "language_loss": 0.78539938,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80751705,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 2.6392276287078857
    },
    {
      "auxiliary_loss_clip": 0.01173938,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.0534054,
      "balance_loss_mlp": 1.01946092,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.9969595373114517,
      "language_loss": 0.64049602,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.66251457,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.6600844860076904
    },
    {
      "auxiliary_loss_clip": 0.01169193,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.05399346,
      "balance_loss_mlp": 1.02504134,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 1.9053512786337807,
      "language_loss": 0.76461816,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.78664911,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.667070150375366
    },
    {
      "auxiliary_loss_clip": 0.01161924,
      "auxiliary_loss_mlp": 0.00901695,
      "balance_loss_clip": 1.05205107,
      "balance_loss_mlp": 1.00108051,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 2.136433143107458,
      "language_loss": 0.75330335,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.77393961,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.7376859188079834
    },
    {
      "auxiliary_loss_clip": 0.01185525,
      "auxiliary_loss_mlp": 0.0102793,
      "balance_loss_clip": 1.05733967,
      "balance_loss_mlp": 1.01991868,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.5660588835703841,
      "language_loss": 0.74943328,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.77156788,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.6335930824279785
    },
    {
      "auxiliary_loss_clip": 0.01195434,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.05801082,
      "balance_loss_mlp": 1.01815104,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.9596890330894727,
      "language_loss": 0.73383462,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75606352,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 3.559854030609131
    },
    {
      "auxiliary_loss_clip": 0.01180261,
      "auxiliary_loss_mlp": 0.01027611,
      "balance_loss_clip": 1.05459785,
      "balance_loss_mlp": 1.01952267,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.598142738540997,
      "language_loss": 0.67769641,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.6997751,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.8832626342773438
    },
    {
      "auxiliary_loss_clip": 0.01189415,
      "auxiliary_loss_mlp": 0.01029127,
      "balance_loss_clip": 1.05791581,
      "balance_loss_mlp": 1.02056205,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.9157434074000794,
      "language_loss": 0.80043554,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82262099,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.5389583110809326
    },
    {
      "auxiliary_loss_clip": 0.01193473,
      "auxiliary_loss_mlp": 0.01030496,
      "balance_loss_clip": 1.05870128,
      "balance_loss_mlp": 1.02200794,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 2.2477541413633397,
      "language_loss": 0.74558723,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76782691,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.6276049613952637
    },
    {
      "auxiliary_loss_clip": 0.01170126,
      "auxiliary_loss_mlp": 0.01029061,
      "balance_loss_clip": 1.05477452,
      "balance_loss_mlp": 1.01990581,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.7106821110844044,
      "language_loss": 0.6369018,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65889364,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.688196897506714
    },
    {
      "auxiliary_loss_clip": 0.01188732,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.05477023,
      "balance_loss_mlp": 1.02098894,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 3.6058906365124668,
      "language_loss": 0.83171868,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85390079,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.6639363765716553
    },
    {
      "auxiliary_loss_clip": 0.01179335,
      "auxiliary_loss_mlp": 0.01026006,
      "balance_loss_clip": 1.05611265,
      "balance_loss_mlp": 1.01798296,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.680637812287883,
      "language_loss": 0.79284883,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81490219,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 3.6093666553497314
    },
    {
      "auxiliary_loss_clip": 0.01184025,
      "auxiliary_loss_mlp": 0.01032131,
      "balance_loss_clip": 1.05498838,
      "balance_loss_mlp": 1.02384591,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 5.23203879468652,
      "language_loss": 0.82523453,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84739614,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.6237218379974365
    },
    {
      "auxiliary_loss_clip": 0.01145072,
      "auxiliary_loss_mlp": 0.01025201,
      "balance_loss_clip": 1.04932022,
      "balance_loss_mlp": 1.01666594,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 1.9513187950342235,
      "language_loss": 0.7440744,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76577711,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.7597851753234863
    },
    {
      "auxiliary_loss_clip": 0.01191589,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.05821776,
      "balance_loss_mlp": 1.02097511,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.8743541140451851,
      "language_loss": 0.71572936,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73793745,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 3.5521252155303955
    },
    {
      "auxiliary_loss_clip": 0.01168825,
      "auxiliary_loss_mlp": 0.01028245,
      "balance_loss_clip": 1.05186903,
      "balance_loss_mlp": 1.02010298,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.7417407428482408,
      "language_loss": 0.79207116,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81404185,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.655642032623291
    },
    {
      "auxiliary_loss_clip": 0.01169575,
      "auxiliary_loss_mlp": 0.01023833,
      "balance_loss_clip": 1.05450892,
      "balance_loss_mlp": 1.0158931,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.6256951351235105,
      "language_loss": 0.81327784,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83521193,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.6402196884155273
    },
    {
      "auxiliary_loss_clip": 0.01141014,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04681182,
      "balance_loss_mlp": 1.01860535,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.6940480704527898,
      "language_loss": 0.7921049,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81378007,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.7219882011413574
    },
    {
      "auxiliary_loss_clip": 0.01157939,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.05075216,
      "balance_loss_mlp": 1.01784468,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.5665341897132046,
      "language_loss": 0.75898784,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78082854,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.730182647705078
    },
    {
      "auxiliary_loss_clip": 0.01190446,
      "auxiliary_loss_mlp": 0.00901781,
      "balance_loss_clip": 1.05577087,
      "balance_loss_mlp": 1.00109458,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 2.804404796698148,
      "language_loss": 0.80535984,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82628214,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 3.4580304622650146
    },
    {
      "auxiliary_loss_clip": 0.0119138,
      "auxiliary_loss_mlp": 0.01027229,
      "balance_loss_clip": 1.05826688,
      "balance_loss_mlp": 1.01891971,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.5657117413053547,
      "language_loss": 0.67618299,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.69836915,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.618884563446045
    },
    {
      "auxiliary_loss_clip": 0.01163042,
      "auxiliary_loss_mlp": 0.0102374,
      "balance_loss_clip": 1.05212092,
      "balance_loss_mlp": 1.01551414,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.885275592674671,
      "language_loss": 0.71428221,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.73615003,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.6643269062042236
    },
    {
      "auxiliary_loss_clip": 0.01163796,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.05388689,
      "balance_loss_mlp": 1.02037692,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.098982899027082,
      "language_loss": 0.84474534,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86666554,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.612802505493164
    },
    {
      "auxiliary_loss_clip": 0.01178935,
      "auxiliary_loss_mlp": 0.01022792,
      "balance_loss_clip": 1.05284214,
      "balance_loss_mlp": 1.01444113,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 1.9742475318047168,
      "language_loss": 0.64769423,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66971153,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.5894570350646973
    },
    {
      "auxiliary_loss_clip": 0.01075039,
      "auxiliary_loss_mlp": 0.01004823,
      "balance_loss_clip": 1.02649117,
      "balance_loss_mlp": 1.00354755,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.8240181358348434,
      "language_loss": 0.56498498,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.5857836,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.1855342388153076
    },
    {
      "auxiliary_loss_clip": 0.01176231,
      "auxiliary_loss_mlp": 0.01030146,
      "balance_loss_clip": 1.05708456,
      "balance_loss_mlp": 1.02111006,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 1.873794916392787,
      "language_loss": 0.73161232,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75367606,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.64976167678833
    },
    {
      "auxiliary_loss_clip": 0.01171855,
      "auxiliary_loss_mlp": 0.01028415,
      "balance_loss_clip": 1.05139828,
      "balance_loss_mlp": 1.01997471,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 1.8595499770826376,
      "language_loss": 0.78834492,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.81034768,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.6759040355682373
    },
    {
      "auxiliary_loss_clip": 0.01178318,
      "auxiliary_loss_mlp": 0.01024619,
      "balance_loss_clip": 1.05352139,
      "balance_loss_mlp": 1.0160718,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 3.073545807305705,
      "language_loss": 0.80427647,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82630581,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 2.654879093170166
    },
    {
      "auxiliary_loss_clip": 0.01178662,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.05564547,
      "balance_loss_mlp": 1.02263737,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 1.756619342410557,
      "language_loss": 0.79985237,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82194561,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.6002695560455322
    },
    {
      "auxiliary_loss_clip": 0.01188751,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.05842733,
      "balance_loss_mlp": 1.01641214,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 1.9031524048110884,
      "language_loss": 0.78870738,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81084555,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.616098165512085
    },
    {
      "auxiliary_loss_clip": 0.01189934,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.0582099,
      "balance_loss_mlp": 1.0200603,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.8601742528401213,
      "language_loss": 0.80344403,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82562572,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.5773003101348877
    },
    {
      "auxiliary_loss_clip": 0.01180158,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.05317795,
      "balance_loss_mlp": 1.01656079,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.6057830258306465,
      "language_loss": 0.80425555,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.8263101,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.67096209526062
    },
    {
      "auxiliary_loss_clip": 0.01195647,
      "auxiliary_loss_mlp": 0.01035489,
      "balance_loss_clip": 1.06205761,
      "balance_loss_mlp": 1.0268048,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.0060521801230444,
      "language_loss": 0.71125597,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73356736,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 3.512179136276245
    },
    {
      "auxiliary_loss_clip": 0.01179466,
      "auxiliary_loss_mlp": 0.01030274,
      "balance_loss_clip": 1.05267203,
      "balance_loss_mlp": 1.02087438,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 3.6946943643207546,
      "language_loss": 0.73455918,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75665665,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.6409225463867188
    },
    {
      "auxiliary_loss_clip": 0.01176674,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.05467808,
      "balance_loss_mlp": 1.02204323,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 2.068789547972266,
      "language_loss": 0.82713568,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84921336,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.6447861194610596
    },
    {
      "auxiliary_loss_clip": 0.01169946,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.05447114,
      "balance_loss_mlp": 1.02058864,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.962936894247819,
      "language_loss": 0.64363086,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66561162,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.6607346534729004
    },
    {
      "auxiliary_loss_clip": 0.01195157,
      "auxiliary_loss_mlp": 0.00902162,
      "balance_loss_clip": 1.06041598,
      "balance_loss_mlp": 1.00102949,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 2.0504179993784666,
      "language_loss": 0.7990104,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.8199836,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.6060543060302734
    },
    {
      "auxiliary_loss_clip": 0.01188318,
      "auxiliary_loss_mlp": 0.01024407,
      "balance_loss_clip": 1.05626464,
      "balance_loss_mlp": 1.01633668,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 1.9273901769402024,
      "language_loss": 0.82849312,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.85062039,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.680276870727539
    },
    {
      "auxiliary_loss_clip": 0.01174525,
      "auxiliary_loss_mlp": 0.01033507,
      "balance_loss_clip": 1.05297673,
      "balance_loss_mlp": 1.02493,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.7303108558028586,
      "language_loss": 0.8670038,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.88908416,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 3.5937271118164062
    },
    {
      "auxiliary_loss_clip": 0.01155073,
      "auxiliary_loss_mlp": 0.00901882,
      "balance_loss_clip": 1.04808784,
      "balance_loss_mlp": 1.0009129,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 1.8214292264655378,
      "language_loss": 0.6487307,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.66930026,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.7603490352630615
    },
    {
      "auxiliary_loss_clip": 0.01184717,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.05455005,
      "balance_loss_mlp": 1.02309036,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 1.7722916340200108,
      "language_loss": 0.80448407,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82664287,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.674701452255249
    },
    {
      "auxiliary_loss_clip": 0.01192169,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 1.05873096,
      "balance_loss_mlp": 1.01911497,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.9287489648917366,
      "language_loss": 0.75867373,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78087056,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 3.5237836837768555
    },
    {
      "auxiliary_loss_clip": 0.01176485,
      "auxiliary_loss_mlp": 0.010272,
      "balance_loss_clip": 1.0508846,
      "balance_loss_mlp": 1.01938009,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.8293787925385718,
      "language_loss": 0.67762721,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.69966406,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.7877376079559326
    },
    {
      "auxiliary_loss_clip": 0.01173946,
      "auxiliary_loss_mlp": 0.01029547,
      "balance_loss_clip": 1.05291367,
      "balance_loss_mlp": 1.0219177,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 2.935692529917849,
      "language_loss": 0.79743052,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.8194654,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.6354193687438965
    },
    {
      "auxiliary_loss_clip": 0.01158939,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 1.05663013,
      "balance_loss_mlp": 1.01875627,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 1.6542003345190273,
      "language_loss": 0.70610559,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72795689,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.8524200916290283
    },
    {
      "auxiliary_loss_clip": 0.01153869,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.05122566,
      "balance_loss_mlp": 1.01882827,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.6605222773706085,
      "language_loss": 0.73549104,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.75730562,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.727158308029175
    },
    {
      "auxiliary_loss_clip": 0.0116539,
      "auxiliary_loss_mlp": 0.00902313,
      "balance_loss_clip": 1.05198419,
      "balance_loss_mlp": 1.00102115,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 2.4073216093655736,
      "language_loss": 0.81203431,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.83271134,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 3.6622302532196045
    },
    {
      "auxiliary_loss_clip": 0.01169082,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.05267572,
      "balance_loss_mlp": 1.01686335,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.1052762612688682,
      "language_loss": 0.8399694,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86191154,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.6344573497772217
    },
    {
      "auxiliary_loss_clip": 0.01165134,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.05072403,
      "balance_loss_mlp": 1.01730585,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.8431003264779457,
      "language_loss": 0.719006,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74091351,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.7308833599090576
    },
    {
      "auxiliary_loss_clip": 0.01067087,
      "auxiliary_loss_mlp": 0.01010602,
      "balance_loss_clip": 1.03320038,
      "balance_loss_mlp": 1.00940418,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7881192330529017,
      "language_loss": 0.63684118,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65761805,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.367591619491577
    },
    {
      "auxiliary_loss_clip": 0.01191154,
      "auxiliary_loss_mlp": 0.01031765,
      "balance_loss_clip": 1.05681729,
      "balance_loss_mlp": 1.02312279,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 1.9306658866737545,
      "language_loss": 0.76482958,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78705871,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.597416877746582
    },
    {
      "auxiliary_loss_clip": 0.01172568,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.05175364,
      "balance_loss_mlp": 1.02253199,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 3.426413608545908,
      "language_loss": 0.82395393,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84599137,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6340363025665283
    },
    {
      "auxiliary_loss_clip": 0.01169852,
      "auxiliary_loss_mlp": 0.01030972,
      "balance_loss_clip": 1.05605972,
      "balance_loss_mlp": 1.02337861,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 2.2157122154020703,
      "language_loss": 0.85974574,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88175398,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.6843369007110596
    },
    {
      "auxiliary_loss_clip": 0.01187973,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 1.05614543,
      "balance_loss_mlp": 1.02417326,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 6.337984218106409,
      "language_loss": 0.78261936,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80482304,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 2.555901050567627
    },
    {
      "auxiliary_loss_clip": 0.01090374,
      "auxiliary_loss_mlp": 0.01001169,
      "balance_loss_clip": 1.0210762,
      "balance_loss_mlp": 0.99960786,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.784171347358018,
      "language_loss": 0.62500644,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64592183,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 3.1108062267303467
    },
    {
      "auxiliary_loss_clip": 0.01171726,
      "auxiliary_loss_mlp": 0.01024573,
      "balance_loss_clip": 1.05371976,
      "balance_loss_mlp": 1.0170753,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 2.797972809540567,
      "language_loss": 0.84912252,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87108546,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.6902081966400146
    },
    {
      "auxiliary_loss_clip": 0.01160104,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 1.04629743,
      "balance_loss_mlp": 1.0183804,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 2.2113009452279897,
      "language_loss": 0.69284993,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71472502,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.6895289421081543
    },
    {
      "auxiliary_loss_clip": 0.01168658,
      "auxiliary_loss_mlp": 0.01030806,
      "balance_loss_clip": 1.05129874,
      "balance_loss_mlp": 1.02245593,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 2.437481602368699,
      "language_loss": 0.82131207,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84330672,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.6680688858032227
    },
    {
      "auxiliary_loss_clip": 0.0118274,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.05583429,
      "balance_loss_mlp": 1.0204742,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.7018249185583643,
      "language_loss": 0.67746449,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.69957697,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 2.6486523151397705
    },
    {
      "auxiliary_loss_clip": 0.01146466,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04930115,
      "balance_loss_mlp": 1.01800394,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 1.9818012480489653,
      "language_loss": 0.72395444,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.74567741,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 3.700791597366333
    },
    {
      "auxiliary_loss_clip": 0.01089651,
      "auxiliary_loss_mlp": 0.01004821,
      "balance_loss_clip": 1.01998091,
      "balance_loss_mlp": 1.00327134,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.729623169260919,
      "language_loss": 0.552872,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57381672,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.2266080379486084
    },
    {
      "auxiliary_loss_clip": 0.0116999,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.05357826,
      "balance_loss_mlp": 1.01670837,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 1.8904379753787202,
      "language_loss": 0.66636223,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68831021,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.666607141494751
    },
    {
      "auxiliary_loss_clip": 0.01183943,
      "auxiliary_loss_mlp": 0.01034081,
      "balance_loss_clip": 1.05748749,
      "balance_loss_mlp": 1.02468157,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.3945351873708205,
      "language_loss": 0.85271871,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87489891,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.6513633728027344
    },
    {
      "auxiliary_loss_clip": 0.01154704,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.05442238,
      "balance_loss_mlp": 1.02436566,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 2.006963963699355,
      "language_loss": 0.67453253,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69641161,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.7998080253601074
    },
    {
      "auxiliary_loss_clip": 0.01170241,
      "auxiliary_loss_mlp": 0.01023618,
      "balance_loss_clip": 1.05327725,
      "balance_loss_mlp": 1.01579738,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 8.417390983623104,
      "language_loss": 0.63870007,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66063869,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 3.598693609237671
    },
    {
      "auxiliary_loss_clip": 0.01163894,
      "auxiliary_loss_mlp": 0.01027388,
      "balance_loss_clip": 1.05052626,
      "balance_loss_mlp": 1.01912045,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 2.0238822845988365,
      "language_loss": 0.71732342,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73923624,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.7013614177703857
    },
    {
      "auxiliary_loss_clip": 0.01180531,
      "auxiliary_loss_mlp": 0.0103287,
      "balance_loss_clip": 1.05459785,
      "balance_loss_mlp": 1.02442992,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.8807613965502405,
      "language_loss": 0.80904472,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.83117867,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.6057114601135254
    },
    {
      "auxiliary_loss_clip": 0.01190875,
      "auxiliary_loss_mlp": 0.01030372,
      "balance_loss_clip": 1.06145871,
      "balance_loss_mlp": 1.02242041,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.9853384051731287,
      "language_loss": 0.72867745,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.7508899,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 3.495208978652954
    },
    {
      "auxiliary_loss_clip": 0.01192715,
      "auxiliary_loss_mlp": 0.01030598,
      "balance_loss_clip": 1.05774307,
      "balance_loss_mlp": 1.02292681,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 3.339863023098906,
      "language_loss": 0.79342353,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.81565666,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.5658750534057617
    },
    {
      "auxiliary_loss_clip": 0.01083233,
      "auxiliary_loss_mlp": 0.0100127,
      "balance_loss_clip": 1.01778555,
      "balance_loss_mlp": 0.99981594,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9780146745162598,
      "language_loss": 0.67437774,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69522274,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.0163846015930176
    },
    {
      "auxiliary_loss_clip": 0.01162451,
      "auxiliary_loss_mlp": 0.01034568,
      "balance_loss_clip": 1.05117178,
      "balance_loss_mlp": 1.02638412,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 3.860653267127626,
      "language_loss": 0.83047962,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.85244977,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.615403175354004
    },
    {
      "auxiliary_loss_clip": 0.01178146,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.0525794,
      "balance_loss_mlp": 1.02105963,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.7728631291165846,
      "language_loss": 0.83428264,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85635662,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.5871832370758057
    },
    {
      "auxiliary_loss_clip": 0.01185997,
      "auxiliary_loss_mlp": 0.01028265,
      "balance_loss_clip": 1.05612183,
      "balance_loss_mlp": 1.02040863,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.9691847546698564,
      "language_loss": 0.77215648,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.79429913,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 3.538912057876587
    },
    {
      "auxiliary_loss_clip": 0.01143385,
      "auxiliary_loss_mlp": 0.00901199,
      "balance_loss_clip": 1.04720998,
      "balance_loss_mlp": 1.00095129,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 1.676613444274672,
      "language_loss": 0.78399563,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80444151,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.74324893951416
    },
    {
      "auxiliary_loss_clip": 0.01147989,
      "auxiliary_loss_mlp": 0.00902548,
      "balance_loss_clip": 1.04969358,
      "balance_loss_mlp": 1.00101256,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.4651936066007358,
      "language_loss": 0.80302733,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.8235327,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.7225868701934814
    },
    {
      "auxiliary_loss_clip": 0.01162223,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.04875064,
      "balance_loss_mlp": 1.01631749,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 2.285158338718188,
      "language_loss": 0.73019731,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.75207245,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.6929819583892822
    },
    {
      "auxiliary_loss_clip": 0.0117443,
      "auxiliary_loss_mlp": 0.01030349,
      "balance_loss_clip": 1.05131817,
      "balance_loss_mlp": 1.02181959,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 2.310678245479891,
      "language_loss": 0.74643677,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.76848459,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.6449825763702393
    },
    {
      "auxiliary_loss_clip": 0.01164062,
      "auxiliary_loss_mlp": 0.01027468,
      "balance_loss_clip": 1.05415678,
      "balance_loss_mlp": 1.0186882,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 2.224067979087107,
      "language_loss": 0.77101612,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79293144,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.618135452270508
    },
    {
      "auxiliary_loss_clip": 0.01076814,
      "auxiliary_loss_mlp": 0.00891312,
      "balance_loss_clip": 1.01853585,
      "balance_loss_mlp": 0.99956799,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7283680347946978,
      "language_loss": 0.60144961,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62113082,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.2081680297851562
    },
    {
      "auxiliary_loss_clip": 0.01176542,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.05227089,
      "balance_loss_mlp": 1.02354872,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 1.972609204061515,
      "language_loss": 0.75590354,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.77798736,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.56502103805542
    },
    {
      "auxiliary_loss_clip": 0.01072086,
      "auxiliary_loss_mlp": 0.0100232,
      "balance_loss_clip": 1.01911032,
      "balance_loss_mlp": 1.00073421,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8068392056243809,
      "language_loss": 0.61042899,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63117301,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 3.228464126586914
    },
    {
      "auxiliary_loss_clip": 0.01142136,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.04970157,
      "balance_loss_mlp": 1.0195303,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.3620369566216186,
      "language_loss": 0.70624208,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72794026,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.724224328994751
    },
    {
      "auxiliary_loss_clip": 0.01163474,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.05466151,
      "balance_loss_mlp": 1.0208478,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.5986051636273557,
      "language_loss": 0.78243279,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80435359,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.729398012161255
    },
    {
      "auxiliary_loss_clip": 0.01188021,
      "auxiliary_loss_mlp": 0.01022672,
      "balance_loss_clip": 1.05510879,
      "balance_loss_mlp": 1.01469421,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 2.255373010007074,
      "language_loss": 0.73855191,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.76065886,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.6211459636688232
    },
    {
      "auxiliary_loss_clip": 0.01178387,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.05515873,
      "balance_loss_mlp": 1.0215199,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 1.8928248214836816,
      "language_loss": 0.78271139,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80479503,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 2.6412060260772705
    },
    {
      "auxiliary_loss_clip": 0.01171726,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.05258906,
      "balance_loss_mlp": 1.02113545,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.6594330657476348,
      "language_loss": 0.82895625,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85095924,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 3.5928404331207275
    },
    {
      "auxiliary_loss_clip": 0.01169305,
      "auxiliary_loss_mlp": 0.00901369,
      "balance_loss_clip": 1.05825567,
      "balance_loss_mlp": 1.00090361,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.6445337283832622,
      "language_loss": 0.7122798,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73298657,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.661034107208252
    },
    {
      "auxiliary_loss_clip": 0.01155754,
      "auxiliary_loss_mlp": 0.01023947,
      "balance_loss_clip": 1.05065942,
      "balance_loss_mlp": 1.01618004,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 2.218444107596486,
      "language_loss": 0.78111875,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80291575,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.6788017749786377
    },
    {
      "auxiliary_loss_clip": 0.01155374,
      "auxiliary_loss_mlp": 0.01020608,
      "balance_loss_clip": 1.04771352,
      "balance_loss_mlp": 1.01288652,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 3.0023665577827012,
      "language_loss": 0.82071704,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84247684,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.63944673538208
    },
    {
      "auxiliary_loss_clip": 0.01146634,
      "auxiliary_loss_mlp": 0.01024736,
      "balance_loss_clip": 1.04947209,
      "balance_loss_mlp": 1.01718962,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.9952027180096406,
      "language_loss": 0.7431134,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76482707,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.740391731262207
    },
    {
      "auxiliary_loss_clip": 0.01160746,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.05102491,
      "balance_loss_mlp": 1.02033985,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 2.50334703066602,
      "language_loss": 0.73066795,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75256181,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.7461037635803223
    },
    {
      "auxiliary_loss_clip": 0.01083042,
      "auxiliary_loss_mlp": 0.01002004,
      "balance_loss_clip": 1.01750255,
      "balance_loss_mlp": 1.00064468,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8929414950346144,
      "language_loss": 0.60834181,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62919223,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 4.12410306930542
    },
    {
      "auxiliary_loss_clip": 0.01164776,
      "auxiliary_loss_mlp": 0.00900734,
      "balance_loss_clip": 1.05111468,
      "balance_loss_mlp": 1.00086772,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.5798181776520612,
      "language_loss": 0.81225348,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.83290857,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.85848331451416
    },
    {
      "auxiliary_loss_clip": 0.01180961,
      "auxiliary_loss_mlp": 0.010294,
      "balance_loss_clip": 1.05429721,
      "balance_loss_mlp": 1.02167487,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.240720687660447,
      "language_loss": 0.82314312,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.84524673,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.612111806869507
    },
    {
      "auxiliary_loss_clip": 0.01163618,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.05158234,
      "balance_loss_mlp": 1.02501082,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 1.9951660723349351,
      "language_loss": 0.73634124,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.7583077,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 3.847245693206787
    },
    {
      "auxiliary_loss_clip": 0.01180306,
      "auxiliary_loss_mlp": 0.00901565,
      "balance_loss_clip": 1.05593348,
      "balance_loss_mlp": 1.00081754,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 2.126876449135473,
      "language_loss": 0.82651973,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84733844,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.673020839691162
    },
    {
      "auxiliary_loss_clip": 0.01152448,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.04815567,
      "balance_loss_mlp": 1.02247345,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 1.9319717227113267,
      "language_loss": 0.80710799,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.8289386,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.6649906635284424
    },
    {
      "auxiliary_loss_clip": 0.01168123,
      "auxiliary_loss_mlp": 0.01025349,
      "balance_loss_clip": 1.05369997,
      "balance_loss_mlp": 1.01685524,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.5598852052977406,
      "language_loss": 0.80088866,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82282335,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.729719400405884
    },
    {
      "auxiliary_loss_clip": 0.01158994,
      "auxiliary_loss_mlp": 0.01027785,
      "balance_loss_clip": 1.04957652,
      "balance_loss_mlp": 1.01869547,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.592543266144658,
      "language_loss": 0.66183603,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68370378,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.630038261413574
    },
    {
      "auxiliary_loss_clip": 0.01184716,
      "auxiliary_loss_mlp": 0.01027116,
      "balance_loss_clip": 1.05519509,
      "balance_loss_mlp": 1.01905727,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 2.294469736222651,
      "language_loss": 0.78881383,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.81093216,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 3.5765185356140137
    },
    {
      "auxiliary_loss_clip": 0.01164524,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05110466,
      "balance_loss_mlp": 1.01815534,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 4.383878699535286,
      "language_loss": 0.67794925,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.6998564,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.6354665756225586
    },
    {
      "auxiliary_loss_clip": 0.01169477,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.05176687,
      "balance_loss_mlp": 1.02176464,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 2.0725436724255433,
      "language_loss": 0.63851815,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66050851,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.648601770401001
    },
    {
      "auxiliary_loss_clip": 0.01189308,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.05582309,
      "balance_loss_mlp": 1.02098942,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.1805458938645383,
      "language_loss": 0.72548461,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74767005,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.675515651702881
    },
    {
      "auxiliary_loss_clip": 0.01150267,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.04869282,
      "balance_loss_mlp": 1.02025306,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 2.159260411228802,
      "language_loss": 0.78124928,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80303359,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.8465752601623535
    },
    {
      "auxiliary_loss_clip": 0.01137942,
      "auxiliary_loss_mlp": 0.01029048,
      "balance_loss_clip": 1.04011822,
      "balance_loss_mlp": 1.02027416,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.0471304569486124,
      "language_loss": 0.71533525,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73700511,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.8744468688964844
    },
    {
      "auxiliary_loss_clip": 0.01165312,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.05229497,
      "balance_loss_mlp": 1.01866817,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.68424730616246,
      "language_loss": 0.83309829,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85501802,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.5953550338745117
    },
    {
      "auxiliary_loss_clip": 0.01186268,
      "auxiliary_loss_mlp": 0.01020707,
      "balance_loss_clip": 1.05349076,
      "balance_loss_mlp": 1.01275539,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.1162504218099145,
      "language_loss": 0.8084991,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83056891,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 2.586232900619507
    },
    {
      "auxiliary_loss_clip": 0.01183926,
      "auxiliary_loss_mlp": 0.01029955,
      "balance_loss_clip": 1.05583215,
      "balance_loss_mlp": 1.02275455,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.4200404707613083,
      "language_loss": 0.75234914,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77448797,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.665928363800049
    },
    {
      "auxiliary_loss_clip": 0.0114461,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.04704952,
      "balance_loss_mlp": 1.01801014,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.3927939837732324,
      "language_loss": 0.78694761,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80865383,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.7433314323425293
    },
    {
      "auxiliary_loss_clip": 0.01155545,
      "auxiliary_loss_mlp": 0.0102919,
      "balance_loss_clip": 1.04562616,
      "balance_loss_mlp": 1.02024341,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.3597448187175947,
      "language_loss": 0.79808539,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.81993276,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 2.6757097244262695
    },
    {
      "auxiliary_loss_clip": 0.01151123,
      "auxiliary_loss_mlp": 0.01028573,
      "balance_loss_clip": 1.04893589,
      "balance_loss_mlp": 1.02078009,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.3523305224308766,
      "language_loss": 0.72108543,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74288237,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 2.712693452835083
    },
    {
      "auxiliary_loss_clip": 0.01185781,
      "auxiliary_loss_mlp": 0.01028039,
      "balance_loss_clip": 1.05692112,
      "balance_loss_mlp": 1.02002811,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.9631624617179595,
      "language_loss": 0.74829721,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77043533,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.760164737701416
    },
    {
      "auxiliary_loss_clip": 0.01168452,
      "auxiliary_loss_mlp": 0.01030721,
      "balance_loss_clip": 1.04809427,
      "balance_loss_mlp": 1.02250159,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 2.099148516238016,
      "language_loss": 0.78407472,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.80606651,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 3.613898754119873
    },
    {
      "auxiliary_loss_clip": 0.01160244,
      "auxiliary_loss_mlp": 0.0102727,
      "balance_loss_clip": 1.05477476,
      "balance_loss_mlp": 1.01944411,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.9227639785351271,
      "language_loss": 0.82138604,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.84326124,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.6825082302093506
    },
    {
      "auxiliary_loss_clip": 0.01175081,
      "auxiliary_loss_mlp": 0.01022458,
      "balance_loss_clip": 1.05260587,
      "balance_loss_mlp": 1.01480472,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.8368163435395743,
      "language_loss": 0.71837682,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.74035221,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.6832785606384277
    },
    {
      "auxiliary_loss_clip": 0.0116981,
      "auxiliary_loss_mlp": 0.0090106,
      "balance_loss_clip": 1.05411363,
      "balance_loss_mlp": 1.00064182,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 2.4838734899515207,
      "language_loss": 0.77955979,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.80026853,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.751774787902832
    },
    {
      "auxiliary_loss_clip": 0.01167587,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.05289257,
      "balance_loss_mlp": 1.01900041,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 3.0766465768001487,
      "language_loss": 0.75437289,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77631199,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.6245977878570557
    },
    {
      "auxiliary_loss_clip": 0.01175678,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.05284941,
      "balance_loss_mlp": 1.02109861,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.8089133113687628,
      "language_loss": 0.77681923,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79887259,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 3.6169328689575195
    },
    {
      "auxiliary_loss_clip": 0.01182033,
      "auxiliary_loss_mlp": 0.01031557,
      "balance_loss_clip": 1.05420947,
      "balance_loss_mlp": 1.02239037,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 4.671528639212803,
      "language_loss": 0.67598683,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.69812274,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.7728168964385986
    },
    {
      "auxiliary_loss_clip": 0.01152878,
      "auxiliary_loss_mlp": 0.00900984,
      "balance_loss_clip": 1.04807961,
      "balance_loss_mlp": 1.00061321,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.6508537974387179,
      "language_loss": 0.78818727,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80872583,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.67502760887146
    },
    {
      "auxiliary_loss_clip": 0.01085645,
      "auxiliary_loss_mlp": 0.01005244,
      "balance_loss_clip": 1.03511405,
      "balance_loss_mlp": 1.00398016,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6936537507445254,
      "language_loss": 0.54142219,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56233108,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 4.264744281768799
    },
    {
      "auxiliary_loss_clip": 0.0117296,
      "auxiliary_loss_mlp": 0.01022967,
      "balance_loss_clip": 1.0523262,
      "balance_loss_mlp": 1.01553452,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.8790611193516948,
      "language_loss": 0.77777803,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79973727,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.6557934284210205
    },
    {
      "auxiliary_loss_clip": 0.01155595,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.04960847,
      "balance_loss_mlp": 1.02057099,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 1.837092394733735,
      "language_loss": 0.76174122,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78357911,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.755549669265747
    },
    {
      "auxiliary_loss_clip": 0.011669,
      "auxiliary_loss_mlp": 0.01022358,
      "balance_loss_clip": 1.05204272,
      "balance_loss_mlp": 1.01447225,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.3641380030837964,
      "language_loss": 0.67988539,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70177794,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.705702781677246
    },
    {
      "auxiliary_loss_clip": 0.01186362,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.05621588,
      "balance_loss_mlp": 1.01995707,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.7455376395099933,
      "language_loss": 0.74572247,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76786065,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.5798914432525635
    },
    {
      "auxiliary_loss_clip": 0.01159332,
      "auxiliary_loss_mlp": 0.01024312,
      "balance_loss_clip": 1.04964995,
      "balance_loss_mlp": 1.01649177,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 1.9544452133137704,
      "language_loss": 0.81844914,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84028554,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 3.6426446437835693
    },
    {
      "auxiliary_loss_clip": 0.01173536,
      "auxiliary_loss_mlp": 0.0102715,
      "balance_loss_clip": 1.04885197,
      "balance_loss_mlp": 1.01845384,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 2.3467916235642323,
      "language_loss": 0.84114528,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86315209,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.6021387577056885
    },
    {
      "auxiliary_loss_clip": 0.01168878,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.05012846,
      "balance_loss_mlp": 1.01822853,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 2.0695926264752966,
      "language_loss": 0.79130638,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81326425,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.653233051300049
    },
    {
      "auxiliary_loss_clip": 0.01167483,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 1.05151486,
      "balance_loss_mlp": 1.02200806,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 5.012921665887417,
      "language_loss": 0.70672911,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72870928,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.677968740463257
    },
    {
      "auxiliary_loss_clip": 0.01165282,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.05030608,
      "balance_loss_mlp": 1.01891255,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.8325642544159377,
      "language_loss": 0.79553956,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81745875,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.6608147621154785
    },
    {
      "auxiliary_loss_clip": 0.01176223,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.05314493,
      "balance_loss_mlp": 1.01795638,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.7575329532668609,
      "language_loss": 0.79563814,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.81765914,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.6727206707000732
    },
    {
      "auxiliary_loss_clip": 0.0107332,
      "auxiliary_loss_mlp": 0.01005771,
      "balance_loss_clip": 1.02019596,
      "balance_loss_mlp": 1.00463223,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8856821542019289,
      "language_loss": 0.5280726,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54886341,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.1088216304779053
    },
    {
      "auxiliary_loss_clip": 0.01136182,
      "auxiliary_loss_mlp": 0.01026866,
      "balance_loss_clip": 1.04618239,
      "balance_loss_mlp": 1.01856279,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 2.225034326226965,
      "language_loss": 0.75677204,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77840251,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 2.7615692615509033
    },
    {
      "auxiliary_loss_clip": 0.01162059,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05278409,
      "balance_loss_mlp": 1.02226377,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 1.9547903625962273,
      "language_loss": 0.87769771,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89961874,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.6413607597351074
    },
    {
      "auxiliary_loss_clip": 0.011801,
      "auxiliary_loss_mlp": 0.01022302,
      "balance_loss_clip": 1.05127692,
      "balance_loss_mlp": 1.0138973,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 2.4351617342269023,
      "language_loss": 0.65543914,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67746305,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.728590488433838
    },
    {
      "auxiliary_loss_clip": 0.01183069,
      "auxiliary_loss_mlp": 0.01023928,
      "balance_loss_clip": 1.05537498,
      "balance_loss_mlp": 1.01604867,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.6690558242423756,
      "language_loss": 0.77724224,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79931223,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 2.608100652694702
    },
    {
      "auxiliary_loss_clip": 0.01182696,
      "auxiliary_loss_mlp": 0.00901636,
      "balance_loss_clip": 1.05537808,
      "balance_loss_mlp": 1.00071454,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.027442520112746,
      "language_loss": 0.8038795,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.82472289,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.7637171745300293
    },
    {
      "auxiliary_loss_clip": 0.01187365,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.05420578,
      "balance_loss_mlp": 1.01731789,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 2.731948804213288,
      "language_loss": 0.62519431,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.64732075,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.544104814529419
    },
    {
      "auxiliary_loss_clip": 0.01145021,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.04692483,
      "balance_loss_mlp": 1.02018523,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 2.8309926349339976,
      "language_loss": 0.65681803,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.6785515,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 3.629437208175659
    },
    {
      "auxiliary_loss_clip": 0.01161509,
      "auxiliary_loss_mlp": 0.01022875,
      "balance_loss_clip": 1.04784966,
      "balance_loss_mlp": 1.01506066,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 1.8619148152251497,
      "language_loss": 0.72660923,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74845302,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.7133872509002686
    },
    {
      "auxiliary_loss_clip": 0.01187868,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.05495715,
      "balance_loss_mlp": 1.01989412,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.2245349295320516,
      "language_loss": 0.82700419,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84916967,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.5132994651794434
    },
    {
      "auxiliary_loss_clip": 0.0118357,
      "auxiliary_loss_mlp": 0.01027728,
      "balance_loss_clip": 1.05476999,
      "balance_loss_mlp": 1.01882291,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 3.0484378691471514,
      "language_loss": 0.79519749,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.81731051,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.5173487663269043
    },
    {
      "auxiliary_loss_clip": 0.01165777,
      "auxiliary_loss_mlp": 0.01021141,
      "balance_loss_clip": 1.05001879,
      "balance_loss_mlp": 1.0135833,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.7202715943459075,
      "language_loss": 0.73620671,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75807583,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.521237850189209
    },
    {
      "auxiliary_loss_clip": 0.01177287,
      "auxiliary_loss_mlp": 0.00900205,
      "balance_loss_clip": 1.05299163,
      "balance_loss_mlp": 1.00071526,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.6000451187289182,
      "language_loss": 0.67370033,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69447517,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.567366123199463
    },
    {
      "auxiliary_loss_clip": 0.01155887,
      "auxiliary_loss_mlp": 0.0102815,
      "balance_loss_clip": 1.04765582,
      "balance_loss_mlp": 1.02022266,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 2.3499177882832547,
      "language_loss": 0.72277641,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.74461681,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 3.4279205799102783
    },
    {
      "auxiliary_loss_clip": 0.01147215,
      "auxiliary_loss_mlp": 0.0102816,
      "balance_loss_clip": 1.0493896,
      "balance_loss_mlp": 1.02036369,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 2.145461180775391,
      "language_loss": 0.79519862,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81695235,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.7015655040740967
    },
    {
      "auxiliary_loss_clip": 0.01066951,
      "auxiliary_loss_mlp": 0.01002697,
      "balance_loss_clip": 1.01860762,
      "balance_loss_mlp": 1.00153434,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8420716101730347,
      "language_loss": 0.57348394,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59418046,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.128528118133545
    },
    {
      "auxiliary_loss_clip": 0.0106141,
      "auxiliary_loss_mlp": 0.01001302,
      "balance_loss_clip": 1.01912045,
      "balance_loss_mlp": 0.99997234,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8269809113502591,
      "language_loss": 0.61932456,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.63995171,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 4.226596355438232
    },
    {
      "auxiliary_loss_clip": 0.01158311,
      "auxiliary_loss_mlp": 0.00901484,
      "balance_loss_clip": 1.04940224,
      "balance_loss_mlp": 1.00078666,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.217181821202364,
      "language_loss": 0.84562826,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.86622632,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.731865406036377
    },
    {
      "auxiliary_loss_clip": 0.01177112,
      "auxiliary_loss_mlp": 0.01027956,
      "balance_loss_clip": 1.05295873,
      "balance_loss_mlp": 1.02020788,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 2.5045793303963344,
      "language_loss": 0.85918999,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88124067,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.6246564388275146
    },
    {
      "auxiliary_loss_clip": 0.01151439,
      "auxiliary_loss_mlp": 0.0090138,
      "balance_loss_clip": 1.05104244,
      "balance_loss_mlp": 1.00061929,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.1095309641366233,
      "language_loss": 0.8030104,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.8235386,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.6687815189361572
    },
    {
      "auxiliary_loss_clip": 0.0114019,
      "auxiliary_loss_mlp": 0.01032942,
      "balance_loss_clip": 1.04704142,
      "balance_loss_mlp": 1.0254972,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.871191441771759,
      "language_loss": 0.82385182,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.8455832,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.7233052253723145
    },
    {
      "auxiliary_loss_clip": 0.01177362,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 1.05084348,
      "balance_loss_mlp": 1.01561356,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.3662156411075674,
      "language_loss": 0.79061389,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81262356,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 3.5528764724731445
    },
    {
      "auxiliary_loss_clip": 0.01171472,
      "auxiliary_loss_mlp": 0.01028337,
      "balance_loss_clip": 1.05153501,
      "balance_loss_mlp": 1.02014709,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.8237942619026186,
      "language_loss": 0.67472047,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.69671857,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.699251890182495
    },
    {
      "auxiliary_loss_clip": 0.01174426,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.05359435,
      "balance_loss_mlp": 1.0160588,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.2218451877445333,
      "language_loss": 0.80128586,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.82327157,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.536327838897705
    },
    {
      "auxiliary_loss_clip": 0.0114296,
      "auxiliary_loss_mlp": 0.01031071,
      "balance_loss_clip": 1.05094528,
      "balance_loss_mlp": 1.02378702,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 1.9771262253662483,
      "language_loss": 0.73739958,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.7591399,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.789144277572632
    },
    {
      "auxiliary_loss_clip": 0.01162924,
      "auxiliary_loss_mlp": 0.00901106,
      "balance_loss_clip": 1.05085421,
      "balance_loss_mlp": 1.00059628,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 1.9414994445038019,
      "language_loss": 0.72467339,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74531364,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.6829020977020264
    },
    {
      "auxiliary_loss_clip": 0.01155535,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.05023682,
      "balance_loss_mlp": 1.02007818,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.133453076001588,
      "language_loss": 0.70138842,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72322816,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.6727406978607178
    },
    {
      "auxiliary_loss_clip": 0.01167455,
      "auxiliary_loss_mlp": 0.0102797,
      "balance_loss_clip": 1.04885483,
      "balance_loss_mlp": 1.02060318,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 1.9471660990573452,
      "language_loss": 0.81543219,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.83738649,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.774381160736084
    },
    {
      "auxiliary_loss_clip": 0.0117049,
      "auxiliary_loss_mlp": 0.01033857,
      "balance_loss_clip": 1.05097675,
      "balance_loss_mlp": 1.02577162,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.9065638369833742,
      "language_loss": 0.74596822,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76801169,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 2.6218924522399902
    },
    {
      "auxiliary_loss_clip": 0.01167675,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.05127954,
      "balance_loss_mlp": 1.01856756,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 1.9417355590065415,
      "language_loss": 0.8693549,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.8912909,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.6234991550445557
    },
    {
      "auxiliary_loss_clip": 0.01167681,
      "auxiliary_loss_mlp": 0.01032441,
      "balance_loss_clip": 1.05341411,
      "balance_loss_mlp": 1.02410793,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 5.412939683909561,
      "language_loss": 0.67958295,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.7015841,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 2.784468173980713
    },
    {
      "auxiliary_loss_clip": 0.01080624,
      "auxiliary_loss_mlp": 0.0100483,
      "balance_loss_clip": 1.01989019,
      "balance_loss_mlp": 1.00357187,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.0793034651453841,
      "language_loss": 0.77323401,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79408854,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.2158501148223877
    },
    {
      "auxiliary_loss_clip": 0.01171161,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04904175,
      "balance_loss_mlp": 1.01911283,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.9540021282618385,
      "language_loss": 0.91126341,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93324757,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.618767261505127
    },
    {
      "auxiliary_loss_clip": 0.01151585,
      "auxiliary_loss_mlp": 0.01033168,
      "balance_loss_clip": 1.04817104,
      "balance_loss_mlp": 1.02467442,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 4.486729400863897,
      "language_loss": 0.79419899,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81604648,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 3.686647891998291
    },
    {
      "auxiliary_loss_clip": 0.01164763,
      "auxiliary_loss_mlp": 0.01030404,
      "balance_loss_clip": 1.04845142,
      "balance_loss_mlp": 1.02216041,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.8472016489484504,
      "language_loss": 0.76894891,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79090059,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.7145118713378906
    },
    {
      "auxiliary_loss_clip": 0.01176791,
      "auxiliary_loss_mlp": 0.01021917,
      "balance_loss_clip": 1.0537324,
      "balance_loss_mlp": 1.01399589,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.150981942836688,
      "language_loss": 0.80698365,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.82897073,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.6192893981933594
    },
    {
      "auxiliary_loss_clip": 0.01165461,
      "auxiliary_loss_mlp": 0.01026065,
      "balance_loss_clip": 1.0518024,
      "balance_loss_mlp": 1.01834083,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.5411746212395254,
      "language_loss": 0.83052278,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85243809,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.760805368423462
    },
    {
      "auxiliary_loss_clip": 0.01174287,
      "auxiliary_loss_mlp": 0.01025209,
      "balance_loss_clip": 1.05195749,
      "balance_loss_mlp": 1.01685214,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 1.8388702177176832,
      "language_loss": 0.6971001,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.71909505,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.628282070159912
    },
    {
      "auxiliary_loss_clip": 0.01151879,
      "auxiliary_loss_mlp": 0.01023634,
      "balance_loss_clip": 1.04436684,
      "balance_loss_mlp": 1.0154444,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.7027853688954266,
      "language_loss": 0.71412766,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73588276,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.7213354110717773
    },
    {
      "auxiliary_loss_clip": 0.01166548,
      "auxiliary_loss_mlp": 0.0102325,
      "balance_loss_clip": 1.05244029,
      "balance_loss_mlp": 1.01578784,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 1.9659912212074055,
      "language_loss": 0.78621876,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80811673,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 3.5761172771453857
    },
    {
      "auxiliary_loss_clip": 0.01158067,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04937804,
      "balance_loss_mlp": 1.01696289,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 1.9177565612926115,
      "language_loss": 0.66967511,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69150048,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.720522165298462
    },
    {
      "auxiliary_loss_clip": 0.01178005,
      "auxiliary_loss_mlp": 0.00901606,
      "balance_loss_clip": 1.05102348,
      "balance_loss_mlp": 1.00071502,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 2.6193454958319,
      "language_loss": 0.75205553,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77285159,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.761627435684204
    },
    {
      "auxiliary_loss_clip": 0.0117411,
      "auxiliary_loss_mlp": 0.00901824,
      "balance_loss_clip": 1.05396879,
      "balance_loss_mlp": 1.00069726,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.7056132640543877,
      "language_loss": 0.78773391,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.80849326,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 3.6487104892730713
    },
    {
      "auxiliary_loss_clip": 0.01163953,
      "auxiliary_loss_mlp": 0.01022209,
      "balance_loss_clip": 1.05189455,
      "balance_loss_mlp": 1.01526475,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.9801622899219258,
      "language_loss": 0.8449173,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86677897,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.704991340637207
    },
    {
      "auxiliary_loss_clip": 0.01186422,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.05409551,
      "balance_loss_mlp": 1.02025986,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 2.210946330060497,
      "language_loss": 0.76846695,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.79061431,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.5940001010894775
    },
    {
      "auxiliary_loss_clip": 0.01186969,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 1.05201602,
      "balance_loss_mlp": 1.01623702,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.815615492026292,
      "language_loss": 0.82283962,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.84495306,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.6552631855010986
    },
    {
      "auxiliary_loss_clip": 0.01160577,
      "auxiliary_loss_mlp": 0.00901799,
      "balance_loss_clip": 1.04864025,
      "balance_loss_mlp": 1.00069654,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.821446379621219,
      "language_loss": 0.61222333,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.63284707,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.592806339263916
    },
    {
      "auxiliary_loss_clip": 0.01179571,
      "auxiliary_loss_mlp": 0.01030297,
      "balance_loss_clip": 1.0519892,
      "balance_loss_mlp": 1.02219963,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 3.237356710277768,
      "language_loss": 0.7229479,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74504656,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 3.5990378856658936
    },
    {
      "auxiliary_loss_clip": 0.0117458,
      "auxiliary_loss_mlp": 0.01025149,
      "balance_loss_clip": 1.05166352,
      "balance_loss_mlp": 1.01708484,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 5.5368065479523665,
      "language_loss": 0.69508725,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71708459,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.608186721801758
    },
    {
      "auxiliary_loss_clip": 0.01174959,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 1.05487037,
      "balance_loss_mlp": 1.02179015,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 2.140403567929051,
      "language_loss": 0.71484411,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73689097,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.6674768924713135
    },
    {
      "auxiliary_loss_clip": 0.01160396,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.04890227,
      "balance_loss_mlp": 1.01656222,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.718129846628051,
      "language_loss": 0.80765551,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.82950068,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.712808609008789
    },
    {
      "auxiliary_loss_clip": 0.01162968,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.0511229,
      "balance_loss_mlp": 1.01694894,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 3.339700057728791,
      "language_loss": 0.81615984,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83804238,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.6989405155181885
    },
    {
      "auxiliary_loss_clip": 0.01163739,
      "auxiliary_loss_mlp": 0.01029701,
      "balance_loss_clip": 1.04738712,
      "balance_loss_mlp": 1.02174401,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.7521816843021458,
      "language_loss": 0.74713916,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76907361,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.7097058296203613
    },
    {
      "auxiliary_loss_clip": 0.01176264,
      "auxiliary_loss_mlp": 0.0102909,
      "balance_loss_clip": 1.05308139,
      "balance_loss_mlp": 1.02100217,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 3.127984562678764,
      "language_loss": 0.72569239,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74774593,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.6124274730682373
    },
    {
      "auxiliary_loss_clip": 0.01157494,
      "auxiliary_loss_mlp": 0.01023171,
      "balance_loss_clip": 1.05124223,
      "balance_loss_mlp": 1.01592946,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.526759826992996,
      "language_loss": 0.86082613,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88263285,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.7149317264556885
    },
    {
      "auxiliary_loss_clip": 0.01175684,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.05410719,
      "balance_loss_mlp": 1.02024698,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 6.87949844722462,
      "language_loss": 0.77964044,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.80168074,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.709265947341919
    },
    {
      "auxiliary_loss_clip": 0.01186972,
      "auxiliary_loss_mlp": 0.01027836,
      "balance_loss_clip": 1.05518568,
      "balance_loss_mlp": 1.02007568,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.3713607996532655,
      "language_loss": 0.71926516,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74141324,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 2.5902442932128906
    },
    {
      "auxiliary_loss_clip": 0.01179927,
      "auxiliary_loss_mlp": 0.01031147,
      "balance_loss_clip": 1.05120039,
      "balance_loss_mlp": 1.02289772,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.6655987619224724,
      "language_loss": 0.86953515,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89164591,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.5829579830169678
    },
    {
      "auxiliary_loss_clip": 0.0116855,
      "auxiliary_loss_mlp": 0.01024891,
      "balance_loss_clip": 1.05138457,
      "balance_loss_mlp": 1.01725614,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.503930807961735,
      "language_loss": 0.76632762,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78826207,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.7038986682891846
    },
    {
      "auxiliary_loss_clip": 0.01179893,
      "auxiliary_loss_mlp": 0.01027545,
      "balance_loss_clip": 1.05888391,
      "balance_loss_mlp": 1.01918268,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.7100518927321802,
      "language_loss": 0.87226176,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.8943361,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.6263208389282227
    },
    {
      "auxiliary_loss_clip": 0.01185631,
      "auxiliary_loss_mlp": 0.01031301,
      "balance_loss_clip": 1.05566216,
      "balance_loss_mlp": 1.02349877,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.5406800491181962,
      "language_loss": 0.71613997,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73830926,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 3.6013998985290527
    },
    {
      "auxiliary_loss_clip": 0.01184857,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.05231929,
      "balance_loss_mlp": 1.01747632,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 1.8699301946502553,
      "language_loss": 0.72760558,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.74971485,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.6536693572998047
    },
    {
      "auxiliary_loss_clip": 0.01152036,
      "auxiliary_loss_mlp": 0.01025906,
      "balance_loss_clip": 1.05318546,
      "balance_loss_mlp": 1.01846719,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 2.029923006672557,
      "language_loss": 0.64951706,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.67129648,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.6928305625915527
    },
    {
      "auxiliary_loss_clip": 0.01174197,
      "auxiliary_loss_mlp": 0.01022146,
      "balance_loss_clip": 1.05311334,
      "balance_loss_mlp": 1.01468349,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 2.0709836861711106,
      "language_loss": 0.75846469,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78042817,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.6097118854522705
    },
    {
      "auxiliary_loss_clip": 0.01189248,
      "auxiliary_loss_mlp": 0.01028682,
      "balance_loss_clip": 1.05709386,
      "balance_loss_mlp": 1.02017593,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 1.992117614987102,
      "language_loss": 0.79991186,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.8220911,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.578233480453491
    },
    {
      "auxiliary_loss_clip": 0.01135431,
      "auxiliary_loss_mlp": 0.00901085,
      "balance_loss_clip": 1.04548883,
      "balance_loss_mlp": 1.00052691,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.143333578545232,
      "language_loss": 0.73560774,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75597286,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 4.022010564804077
    },
    {
      "auxiliary_loss_clip": 0.0117156,
      "auxiliary_loss_mlp": 0.01033088,
      "balance_loss_clip": 1.05326045,
      "balance_loss_mlp": 1.02495837,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.7719763939637676,
      "language_loss": 0.80190295,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.8239494,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.722700357437134
    },
    {
      "auxiliary_loss_clip": 0.01166279,
      "auxiliary_loss_mlp": 0.01024089,
      "balance_loss_clip": 1.04857635,
      "balance_loss_mlp": 1.01625681,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 2.10187997383636,
      "language_loss": 0.75249612,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77439988,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.6805474758148193
    },
    {
      "auxiliary_loss_clip": 0.01073251,
      "auxiliary_loss_mlp": 0.00891111,
      "balance_loss_clip": 1.02136481,
      "balance_loss_mlp": 0.99967462,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.894423887231633,
      "language_loss": 0.63135529,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.65099895,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 4.252092361450195
    },
    {
      "auxiliary_loss_clip": 0.01184882,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.05459785,
      "balance_loss_mlp": 1.02181172,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.6520261346874565,
      "language_loss": 0.70982337,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73196417,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.649380683898926
    },
    {
      "auxiliary_loss_clip": 0.01165972,
      "auxiliary_loss_mlp": 0.01023107,
      "balance_loss_clip": 1.04919004,
      "balance_loss_mlp": 1.01546025,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 2.2431211048132167,
      "language_loss": 0.80048227,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82237303,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.7014620304107666
    },
    {
      "auxiliary_loss_clip": 0.01148157,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.04939151,
      "balance_loss_mlp": 1.02344429,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.6224080816049216,
      "language_loss": 0.82922447,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85102385,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.8332114219665527
    },
    {
      "auxiliary_loss_clip": 0.011632,
      "auxiliary_loss_mlp": 0.01023155,
      "balance_loss_clip": 1.05040359,
      "balance_loss_mlp": 1.01528716,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.214011397348417,
      "language_loss": 0.76202381,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78388733,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.645920991897583
    },
    {
      "auxiliary_loss_clip": 0.01186576,
      "auxiliary_loss_mlp": 0.00899983,
      "balance_loss_clip": 1.05933356,
      "balance_loss_mlp": 1.00053585,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.807255171550195,
      "language_loss": 0.88114619,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90201175,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.6074209213256836
    },
    {
      "auxiliary_loss_clip": 0.01160106,
      "auxiliary_loss_mlp": 0.01024769,
      "balance_loss_clip": 1.05211496,
      "balance_loss_mlp": 1.0177567,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 2.081180868744091,
      "language_loss": 0.78297353,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80482233,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 3.659595012664795
    },
    {
      "auxiliary_loss_clip": 0.01182172,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.05437756,
      "balance_loss_mlp": 1.0194943,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 2.091545834533264,
      "language_loss": 0.68825507,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71034992,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6227974891662598
    },
    {
      "auxiliary_loss_clip": 0.01173895,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.05063605,
      "balance_loss_mlp": 1.02095211,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.4753028842429958,
      "language_loss": 0.79094255,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81297135,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.7219133377075195
    },
    {
      "auxiliary_loss_clip": 0.01070843,
      "auxiliary_loss_mlp": 0.01002222,
      "balance_loss_clip": 1.01971292,
      "balance_loss_mlp": 1.00081539,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7837587211664762,
      "language_loss": 0.59896982,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.61970055,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.2712833881378174
    },
    {
      "auxiliary_loss_clip": 0.0116215,
      "auxiliary_loss_mlp": 0.0102587,
      "balance_loss_clip": 1.04912555,
      "balance_loss_mlp": 1.01805854,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.0438021963753767,
      "language_loss": 0.82525134,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84713155,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.7545113563537598
    },
    {
      "auxiliary_loss_clip": 0.01175007,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.05142784,
      "balance_loss_mlp": 1.01861858,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 2.612417977767655,
      "language_loss": 0.82664317,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84866595,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.635136604309082
    },
    {
      "auxiliary_loss_clip": 0.01159475,
      "auxiliary_loss_mlp": 0.01025698,
      "balance_loss_clip": 1.04834223,
      "balance_loss_mlp": 1.01805925,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.8279606385275462,
      "language_loss": 0.70463014,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72648185,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.8876404762268066
    },
    {
      "auxiliary_loss_clip": 0.01150597,
      "auxiliary_loss_mlp": 0.01026892,
      "balance_loss_clip": 1.04862475,
      "balance_loss_mlp": 1.01977539,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.8456145200837057,
      "language_loss": 0.71456754,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73634243,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.743685245513916
    },
    {
      "auxiliary_loss_clip": 0.01173143,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.05035746,
      "balance_loss_mlp": 1.02263832,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.7643402704796838,
      "language_loss": 0.74134362,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.76337504,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 2.723607063293457
    },
    {
      "auxiliary_loss_clip": 0.01143645,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.04635358,
      "balance_loss_mlp": 1.02184868,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 2.472503576770641,
      "language_loss": 0.73035824,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75209105,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.7487523555755615
    },
    {
      "auxiliary_loss_clip": 0.01170922,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.04674983,
      "balance_loss_mlp": 1.0169034,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 3.4564539008813293,
      "language_loss": 0.87320554,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89516604,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.708203077316284
    },
    {
      "auxiliary_loss_clip": 0.01147639,
      "auxiliary_loss_mlp": 0.01021483,
      "balance_loss_clip": 1.04853129,
      "balance_loss_mlp": 1.01371098,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.3528437511392455,
      "language_loss": 0.5834254,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60511661,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.7754416465759277
    },
    {
      "auxiliary_loss_clip": 0.01158192,
      "auxiliary_loss_mlp": 0.0102709,
      "balance_loss_clip": 1.04798007,
      "balance_loss_mlp": 1.01912034,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 2.3584772165848857,
      "language_loss": 0.83033562,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85218841,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 3.6553335189819336
    },
    {
      "auxiliary_loss_clip": 0.01146977,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 1.04900932,
      "balance_loss_mlp": 1.02354956,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 3.1832480342595897,
      "language_loss": 0.78658289,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.80837333,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.764730453491211
    },
    {
      "auxiliary_loss_clip": 0.0118927,
      "auxiliary_loss_mlp": 0.0103383,
      "balance_loss_clip": 1.05699635,
      "balance_loss_mlp": 1.02554536,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.100826533462621,
      "language_loss": 0.85115147,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.87338251,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.5755221843719482
    },
    {
      "auxiliary_loss_clip": 0.011822,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.05293632,
      "balance_loss_mlp": 1.02117085,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 1.9328429549954518,
      "language_loss": 0.73854303,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.76065683,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.641313076019287
    },
    {
      "auxiliary_loss_clip": 0.01140952,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.0469439,
      "balance_loss_mlp": 1.01755571,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.7320325135179506,
      "language_loss": 0.74502015,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.76668668,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6897923946380615
    },
    {
      "auxiliary_loss_clip": 0.01082699,
      "auxiliary_loss_mlp": 0.01000823,
      "balance_loss_clip": 1.01809216,
      "balance_loss_mlp": 0.99963665,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8553552538379865,
      "language_loss": 0.61965299,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64048815,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.291048765182495
    },
    {
      "auxiliary_loss_clip": 0.01187256,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.0559206,
      "balance_loss_mlp": 1.01863515,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.728757530864357,
      "language_loss": 0.73340011,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75553238,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 3.5621156692504883
    },
    {
      "auxiliary_loss_clip": 0.01163868,
      "auxiliary_loss_mlp": 0.01032243,
      "balance_loss_clip": 1.05187476,
      "balance_loss_mlp": 1.02511978,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.996566057313792,
      "language_loss": 0.80729198,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.82925308,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.7216968536376953
    },
    {
      "auxiliary_loss_clip": 0.01160058,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.05010462,
      "balance_loss_mlp": 1.02172804,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 1.8943240055104478,
      "language_loss": 0.80042636,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82232398,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.866797685623169
    },
    {
      "auxiliary_loss_clip": 0.01172684,
      "auxiliary_loss_mlp": 0.01025964,
      "balance_loss_clip": 1.04999495,
      "balance_loss_mlp": 1.0181495,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.8040539338724588,
      "language_loss": 0.73242468,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75441116,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 3.651580333709717
    },
    {
      "auxiliary_loss_clip": 0.01158369,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.05033183,
      "balance_loss_mlp": 1.01954138,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 2.118877509618435,
      "language_loss": 0.7613889,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78324723,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.6865038871765137
    },
    {
      "auxiliary_loss_clip": 0.01165652,
      "auxiliary_loss_mlp": 0.00901814,
      "balance_loss_clip": 1.04485846,
      "balance_loss_mlp": 1.00065184,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.848908219357871,
      "language_loss": 0.78060943,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.80128407,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.720055103302002
    },
    {
      "auxiliary_loss_clip": 0.0118601,
      "auxiliary_loss_mlp": 0.01026139,
      "balance_loss_clip": 1.05556107,
      "balance_loss_mlp": 1.01830125,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.0227913536530293,
      "language_loss": 0.78231812,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.8044396,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.5863242149353027
    },
    {
      "auxiliary_loss_clip": 0.01168971,
      "auxiliary_loss_mlp": 0.01024374,
      "balance_loss_clip": 1.05099261,
      "balance_loss_mlp": 1.0168221,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 27.48869092417377,
      "language_loss": 0.82323217,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84516567,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.6884541511535645
    },
    {
      "auxiliary_loss_clip": 0.01170517,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.05445516,
      "balance_loss_mlp": 1.02042913,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.6207170526247145,
      "language_loss": 0.78894359,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81092459,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.6292123794555664
    },
    {
      "auxiliary_loss_clip": 0.0118162,
      "auxiliary_loss_mlp": 0.01023176,
      "balance_loss_clip": 1.0527879,
      "balance_loss_mlp": 1.01520097,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.2856549426243937,
      "language_loss": 0.85408556,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.8761335,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 3.585536003112793
    },
    {
      "auxiliary_loss_clip": 0.01180161,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.05371451,
      "balance_loss_mlp": 1.02247703,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.6254882083652538,
      "language_loss": 0.59154922,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61365962,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.6361520290374756
    },
    {
      "auxiliary_loss_clip": 0.01072458,
      "auxiliary_loss_mlp": 0.01002106,
      "balance_loss_clip": 1.01708865,
      "balance_loss_mlp": 1.00082481,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.7935551155246033,
      "language_loss": 0.60072106,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62146676,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.2021560668945312
    },
    {
      "auxiliary_loss_clip": 0.01162878,
      "auxiliary_loss_mlp": 0.01029603,
      "balance_loss_clip": 1.04756188,
      "balance_loss_mlp": 1.02080834,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 2.5773217778246877,
      "language_loss": 0.76361179,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78553665,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.731797218322754
    },
    {
      "auxiliary_loss_clip": 0.01162874,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.04981732,
      "balance_loss_mlp": 1.01817715,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 2.2230301146768277,
      "language_loss": 0.71563309,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73752403,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.7517130374908447
    },
    {
      "auxiliary_loss_clip": 0.01169548,
      "auxiliary_loss_mlp": 0.01022505,
      "balance_loss_clip": 1.04763031,
      "balance_loss_mlp": 1.01409483,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 2.0475613534468877,
      "language_loss": 0.7095567,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73147726,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.6973869800567627
    },
    {
      "auxiliary_loss_clip": 0.01086251,
      "auxiliary_loss_mlp": 0.01001642,
      "balance_loss_clip": 1.01787639,
      "balance_loss_mlp": 1.00034857,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6885836660798663,
      "language_loss": 0.63783884,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65871781,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.2776401042938232
    },
    {
      "auxiliary_loss_clip": 0.01174714,
      "auxiliary_loss_mlp": 0.01022989,
      "balance_loss_clip": 1.05524838,
      "balance_loss_mlp": 1.01505566,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.552227950062685,
      "language_loss": 0.80662543,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.82860243,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 2.7166810035705566
    },
    {
      "auxiliary_loss_clip": 0.01145463,
      "auxiliary_loss_mlp": 0.01029198,
      "balance_loss_clip": 1.048177,
      "balance_loss_mlp": 1.02162266,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.9249732642371438,
      "language_loss": 0.61832368,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.64007026,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.7582342624664307
    },
    {
      "auxiliary_loss_clip": 0.01144852,
      "auxiliary_loss_mlp": 0.01036213,
      "balance_loss_clip": 1.04622293,
      "balance_loss_mlp": 1.0269742,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.033151733919983,
      "language_loss": 0.82079387,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.84260446,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.7359771728515625
    },
    {
      "auxiliary_loss_clip": 0.0114888,
      "auxiliary_loss_mlp": 0.01032573,
      "balance_loss_clip": 1.04847074,
      "balance_loss_mlp": 1.02469373,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 1.9929715745320475,
      "language_loss": 0.77466035,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79647493,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.729727268218994
    },
    {
      "auxiliary_loss_clip": 0.0117848,
      "auxiliary_loss_mlp": 0.01029605,
      "balance_loss_clip": 1.05339944,
      "balance_loss_mlp": 1.02180588,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.5760769253059221,
      "language_loss": 0.77619147,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79827237,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.6505022048950195
    },
    {
      "auxiliary_loss_clip": 0.01085612,
      "auxiliary_loss_mlp": 0.01001144,
      "balance_loss_clip": 1.01685798,
      "balance_loss_mlp": 0.99988002,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8367622715012644,
      "language_loss": 0.53737384,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55824137,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 4.1831135749816895
    },
    {
      "auxiliary_loss_clip": 0.01085443,
      "auxiliary_loss_mlp": 0.01001078,
      "balance_loss_clip": 1.01711798,
      "balance_loss_mlp": 0.99979013,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8509197975913484,
      "language_loss": 0.63298213,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65384734,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.1867477893829346
    },
    {
      "auxiliary_loss_clip": 0.01174219,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.0509727,
      "balance_loss_mlp": 1.01951909,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 1.8353792901427923,
      "language_loss": 0.79711199,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.81913406,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.6323447227478027
    },
    {
      "auxiliary_loss_clip": 0.01172687,
      "auxiliary_loss_mlp": 0.01029745,
      "balance_loss_clip": 1.05501103,
      "balance_loss_mlp": 1.02168667,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 1.8395828463571164,
      "language_loss": 0.84250534,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86452967,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.696658134460449
    },
    {
      "auxiliary_loss_clip": 0.01150652,
      "auxiliary_loss_mlp": 0.0102953,
      "balance_loss_clip": 1.05288613,
      "balance_loss_mlp": 1.02135837,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.7571853825324304,
      "language_loss": 0.73258007,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.7543819,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.6974728107452393
    },
    {
      "auxiliary_loss_clip": 0.01182525,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.053967,
      "balance_loss_mlp": 1.02015114,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 2.114392016975878,
      "language_loss": 0.79074752,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81286001,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.740544557571411
    },
    {
      "auxiliary_loss_clip": 0.01118138,
      "auxiliary_loss_mlp": 0.01030462,
      "balance_loss_clip": 1.04312062,
      "balance_loss_mlp": 1.022493,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 2.1447668809886724,
      "language_loss": 0.84169495,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86318099,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 3.7944753170013428
    },
    {
      "auxiliary_loss_clip": 0.01151606,
      "auxiliary_loss_mlp": 0.01029778,
      "balance_loss_clip": 1.04660022,
      "balance_loss_mlp": 1.02230918,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 1.9182420636003648,
      "language_loss": 0.82390481,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84571862,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.6501803398132324
    },
    {
      "auxiliary_loss_clip": 0.01187203,
      "auxiliary_loss_mlp": 0.01026255,
      "balance_loss_clip": 1.05564427,
      "balance_loss_mlp": 1.01800525,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.7820305982502949,
      "language_loss": 0.75301832,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77515292,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.678147077560425
    },
    {
      "auxiliary_loss_clip": 0.01174067,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.05350578,
      "balance_loss_mlp": 1.01775646,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.749612265858668,
      "language_loss": 0.79006386,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81206369,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 3.557025909423828
    },
    {
      "auxiliary_loss_clip": 0.01185362,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.05291712,
      "balance_loss_mlp": 1.01752353,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.6623793281932309,
      "language_loss": 0.85080671,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87291849,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.6258628368377686
    },
    {
      "auxiliary_loss_clip": 0.01151938,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.04947209,
      "balance_loss_mlp": 1.01798964,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 1.9863142396994165,
      "language_loss": 0.81246018,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83423805,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.714404344558716
    },
    {
      "auxiliary_loss_clip": 0.01174459,
      "auxiliary_loss_mlp": 0.01024019,
      "balance_loss_clip": 1.0508064,
      "balance_loss_mlp": 1.01637173,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 2.7813888009884837,
      "language_loss": 0.72806561,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75005043,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.6559901237487793
    },
    {
      "auxiliary_loss_clip": 0.01076163,
      "auxiliary_loss_mlp": 0.01005688,
      "balance_loss_clip": 1.02637339,
      "balance_loss_mlp": 1.00446057,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8379748009157993,
      "language_loss": 0.62442964,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64524817,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.225640058517456
    },
    {
      "auxiliary_loss_clip": 0.01110276,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.04115725,
      "balance_loss_mlp": 1.02019525,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 1.8646186686113488,
      "language_loss": 0.68710244,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70848536,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.9155781269073486
    },
    {
      "auxiliary_loss_clip": 0.01083452,
      "auxiliary_loss_mlp": 0.01001275,
      "balance_loss_clip": 1.01542997,
      "balance_loss_mlp": 1.00002337,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.9039156896304801,
      "language_loss": 0.60687339,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62772065,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 4.329995632171631
    },
    {
      "auxiliary_loss_clip": 0.01162193,
      "auxiliary_loss_mlp": 0.01024524,
      "balance_loss_clip": 1.04630172,
      "balance_loss_mlp": 1.01677525,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.3426542745570993,
      "language_loss": 0.85430408,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87617123,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.702117443084717
    },
    {
      "auxiliary_loss_clip": 0.01177328,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.05448747,
      "balance_loss_mlp": 1.02405643,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 2.080788602677651,
      "language_loss": 0.78402293,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80611837,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.6678740978240967
    },
    {
      "auxiliary_loss_clip": 0.01175534,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.05259871,
      "balance_loss_mlp": 1.01670575,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.9289918099124723,
      "language_loss": 0.76823056,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79023063,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.690235137939453
    },
    {
      "auxiliary_loss_clip": 0.01177623,
      "auxiliary_loss_mlp": 0.00901078,
      "balance_loss_clip": 1.05165768,
      "balance_loss_mlp": 1.000579,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 2.3544195318818657,
      "language_loss": 0.78973657,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81052363,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.660677194595337
    },
    {
      "auxiliary_loss_clip": 0.01161669,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.04827905,
      "balance_loss_mlp": 1.01885366,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 3.2644907193507158,
      "language_loss": 0.73477441,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75666177,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.679595708847046
    },
    {
      "auxiliary_loss_clip": 0.0118072,
      "auxiliary_loss_mlp": 0.01028274,
      "balance_loss_clip": 1.0553081,
      "balance_loss_mlp": 1.02051318,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.5976414444239015,
      "language_loss": 0.70953053,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73162049,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 2.6648170948028564
    },
    {
      "auxiliary_loss_clip": 0.01156039,
      "auxiliary_loss_mlp": 0.01033585,
      "balance_loss_clip": 1.04660213,
      "balance_loss_mlp": 1.02560389,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.7891618505112197,
      "language_loss": 0.75499588,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77689219,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.6994810104370117
    },
    {
      "auxiliary_loss_clip": 0.01160034,
      "auxiliary_loss_mlp": 0.0102934,
      "balance_loss_clip": 1.04688597,
      "balance_loss_mlp": 1.02173197,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 1.9995872245720225,
      "language_loss": 0.66137874,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68327248,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.700662851333618
    },
    {
      "auxiliary_loss_clip": 0.011123,
      "auxiliary_loss_mlp": 0.01029524,
      "balance_loss_clip": 1.04125881,
      "balance_loss_mlp": 1.02078629,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.523978254637747,
      "language_loss": 0.66105002,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.6824683,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 3.0009961128234863
    },
    {
      "auxiliary_loss_clip": 0.01177562,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.05047083,
      "balance_loss_mlp": 1.02368474,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.5253286960152044,
      "language_loss": 0.75788575,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.77998447,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 3.4337852001190186
    },
    {
      "auxiliary_loss_clip": 0.01166674,
      "auxiliary_loss_mlp": 0.01031033,
      "balance_loss_clip": 1.0500257,
      "balance_loss_mlp": 1.02262259,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 2.1772685691579414,
      "language_loss": 0.79188037,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81385744,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 3.832717180252075
    },
    {
      "auxiliary_loss_clip": 0.011869,
      "auxiliary_loss_mlp": 0.01033332,
      "balance_loss_clip": 1.05505228,
      "balance_loss_mlp": 1.02496958,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.9510937763594463,
      "language_loss": 0.66544968,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68765205,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.766529083251953
    },
    {
      "auxiliary_loss_clip": 0.01180455,
      "auxiliary_loss_mlp": 0.01029515,
      "balance_loss_clip": 1.05476546,
      "balance_loss_mlp": 1.02109325,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 2.0414906777181767,
      "language_loss": 0.67626464,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.69836432,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.7074639797210693
    },
    {
      "auxiliary_loss_clip": 0.01161753,
      "auxiliary_loss_mlp": 0.01027742,
      "balance_loss_clip": 1.04978681,
      "balance_loss_mlp": 1.01978445,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 2.1391377729197822,
      "language_loss": 0.68769002,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.70958495,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.650285243988037
    },
    {
      "auxiliary_loss_clip": 0.01171666,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.04781044,
      "balance_loss_mlp": 1.01819086,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.942092671487446,
      "language_loss": 0.68516177,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.707142,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.6057727336883545
    },
    {
      "auxiliary_loss_clip": 0.01159543,
      "auxiliary_loss_mlp": 0.01028156,
      "balance_loss_clip": 1.04590178,
      "balance_loss_mlp": 1.02041972,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.9588685875646625,
      "language_loss": 0.65397799,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67585498,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.6679506301879883
    },
    {
      "auxiliary_loss_clip": 0.01186959,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.0540235,
      "balance_loss_mlp": 1.02510059,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 1.900757230536133,
      "language_loss": 0.75609207,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77829689,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.581897258758545
    },
    {
      "auxiliary_loss_clip": 0.01142211,
      "auxiliary_loss_mlp": 0.01034687,
      "balance_loss_clip": 1.04485393,
      "balance_loss_mlp": 1.0261035,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 2.00134087591463,
      "language_loss": 0.76837301,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79014194,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 3.6250524520874023
    },
    {
      "auxiliary_loss_clip": 0.01166688,
      "auxiliary_loss_mlp": 0.01022451,
      "balance_loss_clip": 1.05301619,
      "balance_loss_mlp": 1.01450884,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 1.8709317109692596,
      "language_loss": 0.83442211,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85631353,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.7183687686920166
    },
    {
      "auxiliary_loss_clip": 0.01178818,
      "auxiliary_loss_mlp": 0.01025119,
      "balance_loss_clip": 1.0533967,
      "balance_loss_mlp": 1.01663709,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.6739131724025373,
      "language_loss": 0.74272811,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76476753,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 3.651566743850708
    },
    {
      "auxiliary_loss_clip": 0.01186275,
      "auxiliary_loss_mlp": 0.01030096,
      "balance_loss_clip": 1.05483484,
      "balance_loss_mlp": 1.02173924,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 2.0340338905559374,
      "language_loss": 0.7329163,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75507998,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.556486129760742
    },
    {
      "auxiliary_loss_clip": 0.01151847,
      "auxiliary_loss_mlp": 0.01024795,
      "balance_loss_clip": 1.04887629,
      "balance_loss_mlp": 1.01724899,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.2686838043756956,
      "language_loss": 0.76853877,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.7903052,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.8146512508392334
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.01032392,
      "balance_loss_clip": 1.05196691,
      "balance_loss_mlp": 1.02433348,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 2.4766366412193004,
      "language_loss": 0.80088216,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82303691,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.5545406341552734
    },
    {
      "auxiliary_loss_clip": 0.01184326,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.05289328,
      "balance_loss_mlp": 1.01920819,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 1.9355795789279537,
      "language_loss": 0.72154427,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.74365878,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.577845811843872
    },
    {
      "auxiliary_loss_clip": 0.01185241,
      "auxiliary_loss_mlp": 0.01033019,
      "balance_loss_clip": 1.0533638,
      "balance_loss_mlp": 1.02522242,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 1.8888341317518724,
      "language_loss": 0.78548843,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80767107,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.517117738723755
    },
    {
      "auxiliary_loss_clip": 0.01162228,
      "auxiliary_loss_mlp": 0.01027878,
      "balance_loss_clip": 1.04979718,
      "balance_loss_mlp": 1.01998675,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.8170893910200163,
      "language_loss": 0.78114301,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80304408,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 3.6786608695983887
    },
    {
      "auxiliary_loss_clip": 0.01176133,
      "auxiliary_loss_mlp": 0.01029085,
      "balance_loss_clip": 1.05275559,
      "balance_loss_mlp": 1.01965547,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 8.068469486800467,
      "language_loss": 0.88622642,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.90827864,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.686382293701172
    },
    {
      "auxiliary_loss_clip": 0.01175621,
      "auxiliary_loss_mlp": 0.00901672,
      "balance_loss_clip": 1.05258548,
      "balance_loss_mlp": 1.0005759,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 2.0362263635935864,
      "language_loss": 0.92653728,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.94731021,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.7105531692504883
    },
    {
      "auxiliary_loss_clip": 0.01153747,
      "auxiliary_loss_mlp": 0.00901615,
      "balance_loss_clip": 1.05056131,
      "balance_loss_mlp": 1.00077009,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 3.5361121478241566,
      "language_loss": 0.72868675,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.7492404,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.7225756645202637
    },
    {
      "auxiliary_loss_clip": 0.01162739,
      "auxiliary_loss_mlp": 0.01023984,
      "balance_loss_clip": 1.04929948,
      "balance_loss_mlp": 1.01621473,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.9895960384901843,
      "language_loss": 0.77005911,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79192626,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 2.6532206535339355
    },
    {
      "auxiliary_loss_clip": 0.01148432,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.0487262,
      "balance_loss_mlp": 1.02065217,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 2.6634725168922313,
      "language_loss": 0.77795517,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.79971933,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.8385534286499023
    },
    {
      "auxiliary_loss_clip": 0.01180989,
      "auxiliary_loss_mlp": 0.00901831,
      "balance_loss_clip": 1.05231023,
      "balance_loss_mlp": 1.00061142,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 1.991155346371202,
      "language_loss": 0.82596385,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.8467921,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.685166597366333
    },
    {
      "auxiliary_loss_clip": 0.01167452,
      "auxiliary_loss_mlp": 0.01028619,
      "balance_loss_clip": 1.0521723,
      "balance_loss_mlp": 1.02106667,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 1.8928986339148957,
      "language_loss": 0.88475382,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90671456,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.6839182376861572
    },
    {
      "auxiliary_loss_clip": 0.01172917,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.05190969,
      "balance_loss_mlp": 1.01814067,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 5.483291067799506,
      "language_loss": 0.76412475,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78612113,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.5567989349365234
    },
    {
      "auxiliary_loss_clip": 0.01070965,
      "auxiliary_loss_mlp": 0.01002664,
      "balance_loss_clip": 1.01750934,
      "balance_loss_mlp": 1.00119162,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8300704197573973,
      "language_loss": 0.6057924,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62652868,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.148933172225952
    },
    {
      "auxiliary_loss_clip": 0.01163687,
      "auxiliary_loss_mlp": 0.01024073,
      "balance_loss_clip": 1.04903638,
      "balance_loss_mlp": 1.01555586,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 1.9681878921520128,
      "language_loss": 0.7392292,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76110685,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.606144666671753
    },
    {
      "auxiliary_loss_clip": 0.01177419,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.05461788,
      "balance_loss_mlp": 1.01877594,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.447046494854263,
      "language_loss": 0.64215904,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66420543,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.625854969024658
    },
    {
      "auxiliary_loss_clip": 0.01177644,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.05258632,
      "balance_loss_mlp": 1.01959157,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 2.0307723589323525,
      "language_loss": 0.76173091,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78378439,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 3.558772325515747
    },
    {
      "auxiliary_loss_clip": 0.01075508,
      "auxiliary_loss_mlp": 0.01003332,
      "balance_loss_clip": 1.0167954,
      "balance_loss_mlp": 1.00184774,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8248990275278189,
      "language_loss": 0.5753237,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59611201,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.1586647033691406
    },
    {
      "auxiliary_loss_clip": 0.01168741,
      "auxiliary_loss_mlp": 0.00900899,
      "balance_loss_clip": 1.05373263,
      "balance_loss_mlp": 1.00053334,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 2.7998890094850877,
      "language_loss": 0.79775238,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.81844878,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.6591806411743164
    },
    {
      "auxiliary_loss_clip": 0.01173521,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.04776692,
      "balance_loss_mlp": 1.02066779,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 2.6697653173859504,
      "language_loss": 0.70313954,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72516334,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.774881362915039
    },
    {
      "auxiliary_loss_clip": 0.01177422,
      "auxiliary_loss_mlp": 0.01027438,
      "balance_loss_clip": 1.0552597,
      "balance_loss_mlp": 1.01966858,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.300912641417516,
      "language_loss": 0.84537494,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86742353,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.6231942176818848
    },
    {
      "auxiliary_loss_clip": 0.01176924,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.05380845,
      "balance_loss_mlp": 1.01627314,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 1.6817476990921434,
      "language_loss": 0.78340995,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80542392,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.5830507278442383
    },
    {
      "auxiliary_loss_clip": 0.01181948,
      "auxiliary_loss_mlp": 0.0103184,
      "balance_loss_clip": 1.05531919,
      "balance_loss_mlp": 1.02362072,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.6216545576760388,
      "language_loss": 0.86186349,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88400137,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 3.5885560512542725
    },
    {
      "auxiliary_loss_clip": 0.0115969,
      "auxiliary_loss_mlp": 0.01031057,
      "balance_loss_clip": 1.04817963,
      "balance_loss_mlp": 1.02180064,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 2.2663437793645502,
      "language_loss": 0.66310978,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68501723,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.661698579788208
    },
    {
      "auxiliary_loss_clip": 0.01082972,
      "auxiliary_loss_mlp": 0.01002469,
      "balance_loss_clip": 1.01807773,
      "balance_loss_mlp": 1.00097895,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.849780782102164,
      "language_loss": 0.5941726,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61502695,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 4.11297607421875
    },
    {
      "auxiliary_loss_clip": 0.01177986,
      "auxiliary_loss_mlp": 0.01025179,
      "balance_loss_clip": 1.05394816,
      "balance_loss_mlp": 1.01702523,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 1.97217113072251,
      "language_loss": 0.76239526,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.78442693,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.6910321712493896
    },
    {
      "auxiliary_loss_clip": 0.01179478,
      "auxiliary_loss_mlp": 0.01033696,
      "balance_loss_clip": 1.05275095,
      "balance_loss_mlp": 1.0258404,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 1.9535453558684304,
      "language_loss": 0.82937688,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85150868,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.67675518989563
    },
    {
      "auxiliary_loss_clip": 0.01185227,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.05305839,
      "balance_loss_mlp": 1.0194329,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.0210502808912167,
      "language_loss": 0.81329972,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83542436,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.5467731952667236
    },
    {
      "auxiliary_loss_clip": 0.01172682,
      "auxiliary_loss_mlp": 0.0102529,
      "balance_loss_clip": 1.05298042,
      "balance_loss_mlp": 1.01763654,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 2.3772079958839774,
      "language_loss": 0.67087722,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69285697,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.641841411590576
    },
    {
      "auxiliary_loss_clip": 0.01183203,
      "auxiliary_loss_mlp": 0.01032895,
      "balance_loss_clip": 1.05129194,
      "balance_loss_mlp": 1.02356708,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.276619701225488,
      "language_loss": 0.75688058,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77904153,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.598379611968994
    },
    {
      "auxiliary_loss_clip": 0.01168259,
      "auxiliary_loss_mlp": 0.01026454,
      "balance_loss_clip": 1.04907119,
      "balance_loss_mlp": 1.01875877,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.9339369928956718,
      "language_loss": 0.62304926,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.6449964,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7820942401885986
    },
    {
      "auxiliary_loss_clip": 0.01144292,
      "auxiliary_loss_mlp": 0.01033501,
      "balance_loss_clip": 1.04965043,
      "balance_loss_mlp": 1.02484655,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 2.291326449353289,
      "language_loss": 0.8425746,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86435258,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 3.7075448036193848
    },
    {
      "auxiliary_loss_clip": 0.01176361,
      "auxiliary_loss_mlp": 0.01025981,
      "balance_loss_clip": 1.05273557,
      "balance_loss_mlp": 1.01767766,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 1.9167367567811964,
      "language_loss": 0.79022789,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81225133,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.688391923904419
    },
    {
      "auxiliary_loss_clip": 0.01175666,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.05261338,
      "balance_loss_mlp": 1.01795971,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 2.0573603994587746,
      "language_loss": 0.81568742,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83770323,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.630507230758667
    },
    {
      "auxiliary_loss_clip": 0.01174373,
      "auxiliary_loss_mlp": 0.01026138,
      "balance_loss_clip": 1.050318,
      "balance_loss_mlp": 1.01774621,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 2.998887152587787,
      "language_loss": 0.75741017,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.77941531,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 2.5748531818389893
    },
    {
      "auxiliary_loss_clip": 0.01162258,
      "auxiliary_loss_mlp": 0.01026953,
      "balance_loss_clip": 1.04979587,
      "balance_loss_mlp": 1.0189898,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 1.842099403401751,
      "language_loss": 0.76084638,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78273851,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.757171392440796
    },
    {
      "auxiliary_loss_clip": 0.01139461,
      "auxiliary_loss_mlp": 0.01033328,
      "balance_loss_clip": 1.04476643,
      "balance_loss_mlp": 1.02453089,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 1.6342889127006208,
      "language_loss": 0.76690286,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.78863072,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.8725626468658447
    },
    {
      "auxiliary_loss_clip": 0.01168409,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.05304718,
      "balance_loss_mlp": 1.02180815,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.6732020470474118,
      "language_loss": 0.8115139,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83350366,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.7701804637908936
    },
    {
      "auxiliary_loss_clip": 0.01067806,
      "auxiliary_loss_mlp": 0.01001504,
      "balance_loss_clip": 1.01570535,
      "balance_loss_mlp": 1.00021696,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8220181119900958,
      "language_loss": 0.58102274,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60171586,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.1842586994171143
    },
    {
      "auxiliary_loss_clip": 0.01162094,
      "auxiliary_loss_mlp": 0.01028876,
      "balance_loss_clip": 1.04777801,
      "balance_loss_mlp": 1.02056086,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.6084042871884647,
      "language_loss": 0.75120759,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.7731173,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.7917349338531494
    },
    {
      "auxiliary_loss_clip": 0.01158901,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.04839456,
      "balance_loss_mlp": 1.01985598,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.9493765057009762,
      "language_loss": 0.77342403,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79529232,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.748500108718872
    },
    {
      "auxiliary_loss_clip": 0.01170303,
      "auxiliary_loss_mlp": 0.01026878,
      "balance_loss_clip": 1.04954016,
      "balance_loss_mlp": 1.01811576,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.0016047018391068,
      "language_loss": 0.79229808,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.8142699,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.626565933227539
    },
    {
      "auxiliary_loss_clip": 0.01161753,
      "auxiliary_loss_mlp": 0.01025739,
      "balance_loss_clip": 1.04805648,
      "balance_loss_mlp": 1.017537,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 2.2859865962157353,
      "language_loss": 0.80203354,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82390845,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.698538064956665
    },
    {
      "auxiliary_loss_clip": 0.01161342,
      "auxiliary_loss_mlp": 0.01027183,
      "balance_loss_clip": 1.05058694,
      "balance_loss_mlp": 1.01920772,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.0886872788741675,
      "language_loss": 0.73345792,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.7553432,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 3.61191987991333
    },
    {
      "auxiliary_loss_clip": 0.01172301,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.05423212,
      "balance_loss_mlp": 1.0255419,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 2.0198556117934134,
      "language_loss": 0.71491396,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73697805,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.7527010440826416
    },
    {
      "auxiliary_loss_clip": 0.01179047,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.04949605,
      "balance_loss_mlp": 1.0183239,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.625500057179272,
      "language_loss": 0.70065922,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.72271645,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.5471701622009277
    },
    {
      "auxiliary_loss_clip": 0.01168518,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.05124915,
      "balance_loss_mlp": 1.02999115,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.7096250101505974,
      "language_loss": 0.76949918,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.79156739,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.6717300415039062
    },
    {
      "auxiliary_loss_clip": 0.01173058,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 1.05457616,
      "balance_loss_mlp": 1.01626277,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 2.120654383826225,
      "language_loss": 0.79936993,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82134867,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.598006010055542
    },
    {
      "auxiliary_loss_clip": 0.01143616,
      "auxiliary_loss_mlp": 0.01022386,
      "balance_loss_clip": 1.0458374,
      "balance_loss_mlp": 1.01397562,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.684564506342532,
      "language_loss": 0.82759523,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.84925526,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.7330236434936523
    },
    {
      "auxiliary_loss_clip": 0.01165277,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.04696393,
      "balance_loss_mlp": 1.01823616,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.379694258301929,
      "language_loss": 0.70799553,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72991335,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 3.652127265930176
    },
    {
      "auxiliary_loss_clip": 0.01158404,
      "auxiliary_loss_mlp": 0.01032112,
      "balance_loss_clip": 1.0502634,
      "balance_loss_mlp": 1.02429223,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.8573555072087284,
      "language_loss": 0.84146309,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.86336821,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.625251054763794
    },
    {
      "auxiliary_loss_clip": 0.01168873,
      "auxiliary_loss_mlp": 0.0102934,
      "balance_loss_clip": 1.05250609,
      "balance_loss_mlp": 1.02128124,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 2.384898658718978,
      "language_loss": 0.84572935,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86771142,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 3.640235424041748
    },
    {
      "auxiliary_loss_clip": 0.01165414,
      "auxiliary_loss_mlp": 0.01032057,
      "balance_loss_clip": 1.05374622,
      "balance_loss_mlp": 1.02365828,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 2.6896098824839982,
      "language_loss": 0.80133086,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82330561,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.7609822750091553
    },
    {
      "auxiliary_loss_clip": 0.01187284,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.05509794,
      "balance_loss_mlp": 1.01851797,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.7129648604136107,
      "language_loss": 0.62681198,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64895618,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.6184096336364746
    },
    {
      "auxiliary_loss_clip": 0.01170599,
      "auxiliary_loss_mlp": 0.01028935,
      "balance_loss_clip": 1.05320406,
      "balance_loss_mlp": 1.02061415,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 2.369474677015541,
      "language_loss": 0.68984342,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71183872,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.641975164413452
    },
    {
      "auxiliary_loss_clip": 0.01145828,
      "auxiliary_loss_mlp": 0.01032839,
      "balance_loss_clip": 1.04767752,
      "balance_loss_mlp": 1.02411926,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 2.3026702036795923,
      "language_loss": 0.74328768,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.76507437,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.7207589149475098
    },
    {
      "auxiliary_loss_clip": 0.01174892,
      "auxiliary_loss_mlp": 0.01029443,
      "balance_loss_clip": 1.04988027,
      "balance_loss_mlp": 1.0206399,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 2.711433883460938,
      "language_loss": 0.78503919,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80708253,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.672800302505493
    },
    {
      "auxiliary_loss_clip": 0.0107943,
      "auxiliary_loss_mlp": 0.01005422,
      "balance_loss_clip": 1.0162847,
      "balance_loss_mlp": 1.00417066,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.703941308294023,
      "language_loss": 0.58956391,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.61041242,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 4.270090818405151
    },
    {
      "auxiliary_loss_clip": 0.01178142,
      "auxiliary_loss_mlp": 0.01030304,
      "balance_loss_clip": 1.05520463,
      "balance_loss_mlp": 1.02210283,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.8075042130762955,
      "language_loss": 0.7473923,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76947677,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.6039910316467285
    },
    {
      "auxiliary_loss_clip": 0.01184767,
      "auxiliary_loss_mlp": 0.01026919,
      "balance_loss_clip": 1.0562458,
      "balance_loss_mlp": 1.01788855,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.019274392728952,
      "language_loss": 0.7438519,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76596874,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.5802295207977295
    },
    {
      "auxiliary_loss_clip": 0.01062586,
      "auxiliary_loss_mlp": 0.00891766,
      "balance_loss_clip": 1.01384473,
      "balance_loss_mlp": 0.99964374,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7593260432973904,
      "language_loss": 0.57936424,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.59890771,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.1338706016540527
    },
    {
      "auxiliary_loss_clip": 0.01178628,
      "auxiliary_loss_mlp": 0.01033157,
      "balance_loss_clip": 1.05415249,
      "balance_loss_mlp": 1.02359092,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 2.4906559086619215,
      "language_loss": 0.78811514,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.81023294,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.60017991065979
    },
    {
      "auxiliary_loss_clip": 0.01164189,
      "auxiliary_loss_mlp": 0.01028555,
      "balance_loss_clip": 1.04658663,
      "balance_loss_mlp": 1.01997793,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.965731144891084,
      "language_loss": 0.79982197,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.82174945,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.6302971839904785
    },
    {
      "auxiliary_loss_clip": 0.0118511,
      "auxiliary_loss_mlp": 0.01031359,
      "balance_loss_clip": 1.05383587,
      "balance_loss_mlp": 1.02281737,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.7445484791021373,
      "language_loss": 0.79004401,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81220871,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.6283552646636963
    },
    {
      "auxiliary_loss_clip": 0.01066178,
      "auxiliary_loss_mlp": 0.01001912,
      "balance_loss_clip": 1.01364434,
      "balance_loss_mlp": 1.00060689,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8517560720112429,
      "language_loss": 0.59959322,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62027413,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.230977773666382
    },
    {
      "auxiliary_loss_clip": 0.01188957,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.05609584,
      "balance_loss_mlp": 1.02110398,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 1.9059773316545716,
      "language_loss": 0.74578267,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.76796508,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.6143221855163574
    },
    {
      "auxiliary_loss_clip": 0.01169772,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.04982209,
      "balance_loss_mlp": 1.01666594,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 2.7803906073477327,
      "language_loss": 0.81707418,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.83901858,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.64917254447937
    },
    {
      "auxiliary_loss_clip": 0.01179741,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.05140138,
      "balance_loss_mlp": 1.01946115,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 2.5694407421165435,
      "language_loss": 0.80893272,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83101428,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.6320838928222656
    },
    {
      "auxiliary_loss_clip": 0.01166106,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.04908717,
      "balance_loss_mlp": 1.02154589,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.964349702834379,
      "language_loss": 0.71535307,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73731434,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.667564630508423
    },
    {
      "auxiliary_loss_clip": 0.01187969,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.05410373,
      "balance_loss_mlp": 1.01795864,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.65221204335481,
      "language_loss": 0.74247569,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76461923,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 3.571113109588623
    },
    {
      "auxiliary_loss_clip": 0.01163537,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.04796195,
      "balance_loss_mlp": 1.02326822,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 1.956677416400956,
      "language_loss": 0.72756135,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.74950856,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.7930855751037598
    },
    {
      "auxiliary_loss_clip": 0.01170843,
      "auxiliary_loss_mlp": 0.01025958,
      "balance_loss_clip": 1.05007517,
      "balance_loss_mlp": 1.01802135,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.096799002313909,
      "language_loss": 0.79968107,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82164907,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.6364643573760986
    },
    {
      "auxiliary_loss_clip": 0.01168127,
      "auxiliary_loss_mlp": 0.01030336,
      "balance_loss_clip": 1.05083108,
      "balance_loss_mlp": 1.02189004,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 1.9439270650348672,
      "language_loss": 0.67979097,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70177567,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.852515697479248
    },
    {
      "auxiliary_loss_clip": 0.01162021,
      "auxiliary_loss_mlp": 0.01029465,
      "balance_loss_clip": 1.05227816,
      "balance_loss_mlp": 1.02105463,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 2.242562389969233,
      "language_loss": 0.6853596,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70727444,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.698859214782715
    },
    {
      "auxiliary_loss_clip": 0.01159122,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.05023551,
      "balance_loss_mlp": 1.01961732,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 2.5749301957445723,
      "language_loss": 0.76922792,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.79110372,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.722360134124756
    },
    {
      "auxiliary_loss_clip": 0.01156226,
      "auxiliary_loss_mlp": 0.01032888,
      "balance_loss_clip": 1.05036616,
      "balance_loss_mlp": 1.02379274,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.838631775964097,
      "language_loss": 0.80483592,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82672703,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.713777780532837
    },
    {
      "auxiliary_loss_clip": 0.01151685,
      "auxiliary_loss_mlp": 0.01023329,
      "balance_loss_clip": 1.05175757,
      "balance_loss_mlp": 1.01524687,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.6113924855418587,
      "language_loss": 0.79221141,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81396157,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 3.669776439666748
    },
    {
      "auxiliary_loss_clip": 0.01069083,
      "auxiliary_loss_mlp": 0.0100322,
      "balance_loss_clip": 1.02650619,
      "balance_loss_mlp": 1.00199842,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.8106597042338785,
      "language_loss": 0.54717773,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56790072,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.3730037212371826
    },
    {
      "auxiliary_loss_clip": 0.01183007,
      "auxiliary_loss_mlp": 0.01026215,
      "balance_loss_clip": 1.05255342,
      "balance_loss_mlp": 1.01738811,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 1.9470515730453515,
      "language_loss": 0.78912807,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.81122029,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 3.88451886177063
    },
    {
      "auxiliary_loss_clip": 0.01144561,
      "auxiliary_loss_mlp": 0.01025781,
      "balance_loss_clip": 1.04081583,
      "balance_loss_mlp": 1.01783538,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.2169334706316746,
      "language_loss": 0.64463353,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66633689,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.7635841369628906
    },
    {
      "auxiliary_loss_clip": 0.01164829,
      "auxiliary_loss_mlp": 0.01032511,
      "balance_loss_clip": 1.05288482,
      "balance_loss_mlp": 1.02441716,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 1.967670668389525,
      "language_loss": 0.78548813,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80746162,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.7101588249206543
    },
    {
      "auxiliary_loss_clip": 0.01192328,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.05822432,
      "balance_loss_mlp": 1.01894307,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 4.103366684077914,
      "language_loss": 0.7534405,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77563679,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.5724780559539795
    },
    {
      "auxiliary_loss_clip": 0.01183927,
      "auxiliary_loss_mlp": 0.01033605,
      "balance_loss_clip": 1.05638695,
      "balance_loss_mlp": 1.02490282,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.4441651623105827,
      "language_loss": 0.82546937,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84764469,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.705124855041504
    },
    {
      "auxiliary_loss_clip": 0.01165319,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.04875278,
      "balance_loss_mlp": 1.01894438,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 2.0173497325119816,
      "language_loss": 0.63805187,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.6599772,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.6343977451324463
    },
    {
      "auxiliary_loss_clip": 0.01156791,
      "auxiliary_loss_mlp": 0.01023595,
      "balance_loss_clip": 1.05137181,
      "balance_loss_mlp": 1.01542926,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 1.9119824019967788,
      "language_loss": 0.78675091,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80855477,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 3.724884510040283
    },
    {
      "auxiliary_loss_clip": 0.01164208,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 1.05377328,
      "balance_loss_mlp": 1.01925266,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 1.9896569319542678,
      "language_loss": 0.87506819,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89698458,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.6415417194366455
    },
    {
      "auxiliary_loss_clip": 0.010748,
      "auxiliary_loss_mlp": 0.01002512,
      "balance_loss_clip": 1.01288211,
      "balance_loss_mlp": 1.00124264,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7223898404338346,
      "language_loss": 0.55836481,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57913792,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.1558034420013428
    },
    {
      "auxiliary_loss_clip": 0.01163764,
      "auxiliary_loss_mlp": 0.01026988,
      "balance_loss_clip": 1.04967523,
      "balance_loss_mlp": 1.01901865,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.5473907855806912,
      "language_loss": 0.72485477,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74676228,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.71315860748291
    },
    {
      "auxiliary_loss_clip": 0.0117227,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.0511117,
      "balance_loss_mlp": 1.02088952,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.852997440619754,
      "language_loss": 0.75360477,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77561593,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.6116766929626465
    },
    {
      "auxiliary_loss_clip": 0.01169206,
      "auxiliary_loss_mlp": 0.01031045,
      "balance_loss_clip": 1.05437553,
      "balance_loss_mlp": 1.02291441,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 2.1863115895424423,
      "language_loss": 0.70465261,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72665513,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.686264991760254
    },
    {
      "auxiliary_loss_clip": 0.01157551,
      "auxiliary_loss_mlp": 0.01028026,
      "balance_loss_clip": 1.04721296,
      "balance_loss_mlp": 1.01997924,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 2.114517701678683,
      "language_loss": 0.76488191,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.78673768,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.719719648361206
    },
    {
      "auxiliary_loss_clip": 0.01175077,
      "auxiliary_loss_mlp": 0.01026533,
      "balance_loss_clip": 1.05333591,
      "balance_loss_mlp": 1.01795602,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 3.2252693625359243,
      "language_loss": 0.69951773,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72153378,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.787747621536255
    },
    {
      "auxiliary_loss_clip": 0.01160044,
      "auxiliary_loss_mlp": 0.01034289,
      "balance_loss_clip": 1.04797983,
      "balance_loss_mlp": 1.02589095,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 2.028658962077015,
      "language_loss": 0.90319037,
      "learning_rate": 2e-06,
      "loss": 0.92513371,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.6872406005859375
    },
    {
      "auxiliary_loss_clip": 0.01186082,
      "auxiliary_loss_mlp": 0.01031233,
      "balance_loss_clip": 1.05584979,
      "balance_loss_mlp": 1.02299547,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 3.7972990981186396,
      "language_loss": 0.85972822,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.88190138,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.6047134399414062
    },
    {
      "auxiliary_loss_clip": 0.01185779,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.05570221,
      "balance_loss_mlp": 1.02386367,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.575398151193112,
      "language_loss": 0.79562151,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81779599,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.7238566875457764
    },
    {
      "auxiliary_loss_clip": 0.01175375,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.05158615,
      "balance_loss_mlp": 1.01921368,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 1.9815241032620752,
      "language_loss": 0.78266072,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80468267,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.653705358505249
    },
    {
      "auxiliary_loss_clip": 0.01160153,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04442918,
      "balance_loss_mlp": 1.01995099,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.0423631429198057,
      "language_loss": 0.74912989,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.77101064,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 3.789970874786377
    },
    {
      "auxiliary_loss_clip": 0.0113868,
      "auxiliary_loss_mlp": 0.0090107,
      "balance_loss_clip": 1.04585528,
      "balance_loss_mlp": 1.00079799,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.609326996836101,
      "language_loss": 0.71493262,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.7353301,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.7757954597473145
    },
    {
      "auxiliary_loss_clip": 0.01178261,
      "auxiliary_loss_mlp": 0.01028945,
      "balance_loss_clip": 1.05490541,
      "balance_loss_mlp": 1.02120817,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 3.054593999923881,
      "language_loss": 0.7900387,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.81211066,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.759702682495117
    },
    {
      "auxiliary_loss_clip": 0.01172602,
      "auxiliary_loss_mlp": 0.00900863,
      "balance_loss_clip": 1.0512532,
      "balance_loss_mlp": 1.0006181,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.6129473241861316,
      "language_loss": 0.73263848,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75337309,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.7679009437561035
    },
    {
      "auxiliary_loss_clip": 0.01159132,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.04927635,
      "balance_loss_mlp": 1.01882315,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 2.2222542863879324,
      "language_loss": 0.6779691,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69983619,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.7144668102264404
    },
    {
      "auxiliary_loss_clip": 0.01162111,
      "auxiliary_loss_mlp": 0.01025746,
      "balance_loss_clip": 1.04859102,
      "balance_loss_mlp": 1.01737809,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 2.4039998660972484,
      "language_loss": 0.65878904,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.68066758,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.9301464557647705
    },
    {
      "auxiliary_loss_clip": 0.01168674,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.05014551,
      "balance_loss_mlp": 1.01975226,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 9.887907715854654,
      "language_loss": 0.68761355,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.70958191,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 3.808256149291992
    },
    {
      "auxiliary_loss_clip": 0.01170219,
      "auxiliary_loss_mlp": 0.01030602,
      "balance_loss_clip": 1.05134702,
      "balance_loss_mlp": 1.02277625,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.6508374637218366,
      "language_loss": 0.80926466,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.8312729,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.7341361045837402
    },
    {
      "auxiliary_loss_clip": 0.01167829,
      "auxiliary_loss_mlp": 0.01030217,
      "balance_loss_clip": 1.0551579,
      "balance_loss_mlp": 1.02288556,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 2.4095239237653607,
      "language_loss": 0.79477465,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81675512,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.790893793106079
    },
    {
      "auxiliary_loss_clip": 0.01185134,
      "auxiliary_loss_mlp": 0.01035519,
      "balance_loss_clip": 1.05184674,
      "balance_loss_mlp": 1.02663231,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.394909143663656,
      "language_loss": 0.77647221,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79867876,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 3.5107972621917725
    },
    {
      "auxiliary_loss_clip": 0.01135266,
      "auxiliary_loss_mlp": 0.01028798,
      "balance_loss_clip": 1.04653549,
      "balance_loss_mlp": 1.02000594,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.5529021162423986,
      "language_loss": 0.80426574,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.8259064,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.848595142364502
    },
    {
      "auxiliary_loss_clip": 0.01176198,
      "auxiliary_loss_mlp": 0.01021131,
      "balance_loss_clip": 1.05361521,
      "balance_loss_mlp": 1.01342392,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 6.388986834798571,
      "language_loss": 0.76419097,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.78616428,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.721798896789551
    },
    {
      "auxiliary_loss_clip": 0.01184799,
      "auxiliary_loss_mlp": 0.01024525,
      "balance_loss_clip": 1.05403757,
      "balance_loss_mlp": 1.01717019,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.7270506959198078,
      "language_loss": 0.83350885,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85560203,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.6258389949798584
    },
    {
      "auxiliary_loss_clip": 0.01175113,
      "auxiliary_loss_mlp": 0.00901312,
      "balance_loss_clip": 1.05443692,
      "balance_loss_mlp": 1.00056291,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.667237090144698,
      "language_loss": 0.84819603,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.86896026,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.5992233753204346
    },
    {
      "auxiliary_loss_clip": 0.01178352,
      "auxiliary_loss_mlp": 0.0102713,
      "balance_loss_clip": 1.05365121,
      "balance_loss_mlp": 1.01858294,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.1129699458240516,
      "language_loss": 0.85731828,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.87937313,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 3.649522066116333
    },
    {
      "auxiliary_loss_clip": 0.01162874,
      "auxiliary_loss_mlp": 0.01028329,
      "balance_loss_clip": 1.05204129,
      "balance_loss_mlp": 1.01963842,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 3.062337892973726,
      "language_loss": 0.66875172,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.69066375,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.849426746368408
    },
    {
      "auxiliary_loss_clip": 0.01180084,
      "auxiliary_loss_mlp": 0.01030088,
      "balance_loss_clip": 1.05354643,
      "balance_loss_mlp": 1.02142191,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.416653524599593,
      "language_loss": 0.74675512,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.76885682,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.6070475578308105
    },
    {
      "auxiliary_loss_clip": 0.01162459,
      "auxiliary_loss_mlp": 0.01024156,
      "balance_loss_clip": 1.04926252,
      "balance_loss_mlp": 1.01669395,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.8122622794462744,
      "language_loss": 0.6849848,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70685101,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.772674560546875
    },
    {
      "auxiliary_loss_clip": 0.01072987,
      "auxiliary_loss_mlp": 0.01000817,
      "balance_loss_clip": 1.01360345,
      "balance_loss_mlp": 0.99947566,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.748940620069101,
      "language_loss": 0.58716822,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60790622,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.3474481105804443
    },
    {
      "auxiliary_loss_clip": 0.01073754,
      "auxiliary_loss_mlp": 0.01002415,
      "balance_loss_clip": 1.01366949,
      "balance_loss_mlp": 1.00128222,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7648293980393626,
      "language_loss": 0.55122054,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.5719822,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.2223801612854004
    },
    {
      "auxiliary_loss_clip": 0.01162612,
      "auxiliary_loss_mlp": 0.01024271,
      "balance_loss_clip": 1.05412674,
      "balance_loss_mlp": 1.01665652,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.5447299075213703,
      "language_loss": 0.8290385,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.85090727,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.7269675731658936
    },
    {
      "auxiliary_loss_clip": 0.01155386,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04703069,
      "balance_loss_mlp": 1.02089262,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.733333145723386,
      "language_loss": 0.82156658,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84340638,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.647676706314087
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.01022717,
      "balance_loss_clip": 1.05492914,
      "balance_loss_mlp": 1.01514137,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 2.5238060191588585,
      "language_loss": 0.86493331,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88692224,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.7054758071899414
    },
    {
      "auxiliary_loss_clip": 0.01177326,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.05451858,
      "balance_loss_mlp": 1.02195501,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 2.087801832752772,
      "language_loss": 0.77699876,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79907769,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.6020307540893555
    },
    {
      "auxiliary_loss_clip": 0.01144998,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.04917455,
      "balance_loss_mlp": 1.01714063,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 2.2745550050094128,
      "language_loss": 0.72806406,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74975562,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.7470738887786865
    },
    {
      "auxiliary_loss_clip": 0.01185486,
      "auxiliary_loss_mlp": 0.01026958,
      "balance_loss_clip": 1.05490518,
      "balance_loss_mlp": 1.01869059,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.5512776743282313,
      "language_loss": 0.76984847,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79197288,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.558589220046997
    },
    {
      "auxiliary_loss_clip": 0.0115291,
      "auxiliary_loss_mlp": 0.01019817,
      "balance_loss_clip": 1.05510712,
      "balance_loss_mlp": 1.01237869,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.5744955060134125,
      "language_loss": 0.76447183,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78619909,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 3.768113851547241
    },
    {
      "auxiliary_loss_clip": 0.01143323,
      "auxiliary_loss_mlp": 0.01028001,
      "balance_loss_clip": 1.04484153,
      "balance_loss_mlp": 1.01959658,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.1728864589231995,
      "language_loss": 0.69238436,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71409756,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.7565457820892334
    },
    {
      "auxiliary_loss_clip": 0.01179077,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.05504656,
      "balance_loss_mlp": 1.0243057,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.1558712445426913,
      "language_loss": 0.70457834,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72669041,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.667489767074585
    },
    {
      "auxiliary_loss_clip": 0.01166411,
      "auxiliary_loss_mlp": 0.01024778,
      "balance_loss_clip": 1.05431223,
      "balance_loss_mlp": 1.0169462,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 3.887373541014305,
      "language_loss": 0.7949301,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81684196,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.709127902984619
    },
    {
      "auxiliary_loss_clip": 0.01179164,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.05186558,
      "balance_loss_mlp": 1.020329,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 2.166935012774557,
      "language_loss": 0.66872478,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69079816,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.672914981842041
    },
    {
      "auxiliary_loss_clip": 0.01169201,
      "auxiliary_loss_mlp": 0.01025457,
      "balance_loss_clip": 1.05141127,
      "balance_loss_mlp": 1.01826334,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.7294704863189276,
      "language_loss": 0.8397752,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86172181,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.7846813201904297
    },
    {
      "auxiliary_loss_clip": 0.01154052,
      "auxiliary_loss_mlp": 0.01026964,
      "balance_loss_clip": 1.04863632,
      "balance_loss_mlp": 1.01911986,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 2.3297636481184383,
      "language_loss": 0.68347275,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70528293,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 3.6865222454071045
    },
    {
      "auxiliary_loss_clip": 0.01188749,
      "auxiliary_loss_mlp": 0.01023934,
      "balance_loss_clip": 1.05861449,
      "balance_loss_mlp": 1.01620901,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 3.0563230141975444,
      "language_loss": 0.93738616,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95951301,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.5786449909210205
    },
    {
      "auxiliary_loss_clip": 0.01155497,
      "auxiliary_loss_mlp": 0.01032147,
      "balance_loss_clip": 1.05175853,
      "balance_loss_mlp": 1.02395678,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 3.252090425296459,
      "language_loss": 0.71414047,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73601687,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.730609655380249
    },
    {
      "auxiliary_loss_clip": 0.01160035,
      "auxiliary_loss_mlp": 0.00901487,
      "balance_loss_clip": 1.04727781,
      "balance_loss_mlp": 1.00089157,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 3.0600070939097352,
      "language_loss": 0.8697502,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.89036542,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 3.6255736351013184
    },
    {
      "auxiliary_loss_clip": 0.01182208,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.05206358,
      "balance_loss_mlp": 1.01908541,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 3.679969614421836,
      "language_loss": 0.82167387,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84377247,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.7203140258789062
    },
    {
      "auxiliary_loss_clip": 0.01137905,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.04128718,
      "balance_loss_mlp": 1.01912427,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.2793365761910183,
      "language_loss": 0.72535515,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74700916,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.7489819526672363
    },
    {
      "auxiliary_loss_clip": 0.01176173,
      "auxiliary_loss_mlp": 0.01035125,
      "balance_loss_clip": 1.05064344,
      "balance_loss_mlp": 1.02645802,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.7453642883662863,
      "language_loss": 0.72728789,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.74940085,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.7152247428894043
    },
    {
      "auxiliary_loss_clip": 0.01153101,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.04652882,
      "balance_loss_mlp": 1.02472126,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 1.8931425714429069,
      "language_loss": 0.72537762,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74723691,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.8731253147125244
    },
    {
      "auxiliary_loss_clip": 0.01159283,
      "auxiliary_loss_mlp": 0.01028012,
      "balance_loss_clip": 1.05275416,
      "balance_loss_mlp": 1.01984024,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.6804725629456034,
      "language_loss": 0.76214087,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78401387,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.0880236625671387
    },
    {
      "auxiliary_loss_clip": 0.01185873,
      "auxiliary_loss_mlp": 0.01024492,
      "balance_loss_clip": 1.05643904,
      "balance_loss_mlp": 1.01726246,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.7664938804011285,
      "language_loss": 0.84532881,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.86743242,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 3.6173017024993896
    },
    {
      "auxiliary_loss_clip": 0.01164178,
      "auxiliary_loss_mlp": 0.01031467,
      "balance_loss_clip": 1.0478282,
      "balance_loss_mlp": 1.02370024,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 3.436220007753603,
      "language_loss": 0.702259,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.72421539,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.6911704540252686
    },
    {
      "auxiliary_loss_clip": 0.01149652,
      "auxiliary_loss_mlp": 0.01029803,
      "balance_loss_clip": 1.0542326,
      "balance_loss_mlp": 1.02143478,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 2.132601784377784,
      "language_loss": 0.72354418,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74533874,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 3.136935234069824
    },
    {
      "auxiliary_loss_clip": 0.01183065,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.05251193,
      "balance_loss_mlp": 1.01794505,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 2.737249335129119,
      "language_loss": 0.75199121,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.77407789,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.6462624073028564
    },
    {
      "auxiliary_loss_clip": 0.01165166,
      "auxiliary_loss_mlp": 0.01024322,
      "balance_loss_clip": 1.05185199,
      "balance_loss_mlp": 1.01629913,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 1.9635697279299549,
      "language_loss": 0.85474551,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87664038,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.738548994064331
    },
    {
      "auxiliary_loss_clip": 0.01152824,
      "auxiliary_loss_mlp": 0.0090163,
      "balance_loss_clip": 1.04882717,
      "balance_loss_mlp": 1.00054491,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 2.183678837894813,
      "language_loss": 0.79020274,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81074727,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.739319086074829
    },
    {
      "auxiliary_loss_clip": 0.0107017,
      "auxiliary_loss_mlp": 0.01003294,
      "balance_loss_clip": 1.01706088,
      "balance_loss_mlp": 1.0020541,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7727760736217746,
      "language_loss": 0.59397793,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.6147126,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.335451126098633
    },
    {
      "auxiliary_loss_clip": 0.01163976,
      "auxiliary_loss_mlp": 0.01021534,
      "balance_loss_clip": 1.05009675,
      "balance_loss_mlp": 1.01389265,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.0336322021530346,
      "language_loss": 0.83129013,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.85314518,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.6597349643707275
    },
    {
      "auxiliary_loss_clip": 0.01182904,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 1.06024146,
      "balance_loss_mlp": 1.02652895,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.1795896243648962,
      "language_loss": 0.79784232,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.82002008,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.641239643096924
    },
    {
      "auxiliary_loss_clip": 0.01131897,
      "auxiliary_loss_mlp": 0.01031004,
      "balance_loss_clip": 1.04317975,
      "balance_loss_mlp": 1.02318954,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.3841259406460305,
      "language_loss": 0.82031679,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84194583,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.7205357551574707
    },
    {
      "auxiliary_loss_clip": 0.0119046,
      "auxiliary_loss_mlp": 0.00901561,
      "balance_loss_clip": 1.05853963,
      "balance_loss_mlp": 1.00061405,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.8718348337128057,
      "language_loss": 0.81361085,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.83453101,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.7255773544311523
    },
    {
      "auxiliary_loss_clip": 0.01151449,
      "auxiliary_loss_mlp": 0.01022092,
      "balance_loss_clip": 1.04654038,
      "balance_loss_mlp": 1.0143609,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.43123357752846,
      "language_loss": 0.6899696,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71170497,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.7639710903167725
    },
    {
      "auxiliary_loss_clip": 0.01153297,
      "auxiliary_loss_mlp": 0.01030543,
      "balance_loss_clip": 1.04938805,
      "balance_loss_mlp": 1.02251995,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.8282632341884404,
      "language_loss": 0.8639515,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88578993,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 3.6301136016845703
    },
    {
      "auxiliary_loss_clip": 0.01160087,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.05062103,
      "balance_loss_mlp": 1.02155328,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.8080150300730964,
      "language_loss": 0.79786432,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.81976044,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.717646837234497
    },
    {
      "auxiliary_loss_clip": 0.01177518,
      "auxiliary_loss_mlp": 0.01030891,
      "balance_loss_clip": 1.05533576,
      "balance_loss_mlp": 1.02326763,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.7330858361023722,
      "language_loss": 0.81394589,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83603001,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.726020574569702
    },
    {
      "auxiliary_loss_clip": 0.01076763,
      "auxiliary_loss_mlp": 0.01005528,
      "balance_loss_clip": 1.01639307,
      "balance_loss_mlp": 1.00417507,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8647331728591277,
      "language_loss": 0.60644668,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62726963,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.389090061187744
    },
    {
      "auxiliary_loss_clip": 0.01176684,
      "auxiliary_loss_mlp": 0.01024523,
      "balance_loss_clip": 1.05263901,
      "balance_loss_mlp": 1.01679826,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 2.0259874209659907,
      "language_loss": 0.68145216,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.70346427,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 2.9936695098876953
    },
    {
      "auxiliary_loss_clip": 0.01177901,
      "auxiliary_loss_mlp": 0.0102674,
      "balance_loss_clip": 1.05475581,
      "balance_loss_mlp": 1.01883662,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.1294557806614205,
      "language_loss": 0.8074559,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.82950234,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.664076328277588
    },
    {
      "auxiliary_loss_clip": 0.01166375,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.04882288,
      "balance_loss_mlp": 1.02072406,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.830285911980552,
      "language_loss": 0.82100904,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84295833,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 3.602513074874878
    },
    {
      "auxiliary_loss_clip": 0.01183537,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.05430222,
      "balance_loss_mlp": 1.01806438,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.6745943281498799,
      "language_loss": 0.79211938,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.8142125,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.617457866668701
    },
    {
      "auxiliary_loss_clip": 0.01182999,
      "auxiliary_loss_mlp": 0.01023292,
      "balance_loss_clip": 1.05353963,
      "balance_loss_mlp": 1.01570415,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.2296053519468835,
      "language_loss": 0.81678021,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.83884323,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.575531005859375
    },
    {
      "auxiliary_loss_clip": 0.01155413,
      "auxiliary_loss_mlp": 0.00900491,
      "balance_loss_clip": 1.04932344,
      "balance_loss_mlp": 1.0006237,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 2.5372414534183476,
      "language_loss": 0.8280285,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.84858751,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 3.567093849182129
    },
    {
      "auxiliary_loss_clip": 0.01159158,
      "auxiliary_loss_mlp": 0.01023668,
      "balance_loss_clip": 1.05133247,
      "balance_loss_mlp": 1.01642561,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.6525139539002436,
      "language_loss": 0.83247745,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85430568,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.679422616958618
    },
    {
      "auxiliary_loss_clip": 0.0118394,
      "auxiliary_loss_mlp": 0.01028951,
      "balance_loss_clip": 1.05648541,
      "balance_loss_mlp": 1.02163708,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 2.0481090062383327,
      "language_loss": 0.83247054,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85459936,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.6128177642822266
    },
    {
      "auxiliary_loss_clip": 0.01159516,
      "auxiliary_loss_mlp": 0.00901055,
      "balance_loss_clip": 1.05007339,
      "balance_loss_mlp": 1.00062561,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.6344683073238606,
      "language_loss": 0.81036091,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.83096671,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.790924072265625
    },
    {
      "auxiliary_loss_clip": 0.01164806,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04969418,
      "balance_loss_mlp": 1.01800084,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 2.2636796200463998,
      "language_loss": 0.81140971,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.83331585,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.5996546745300293
    },
    {
      "auxiliary_loss_clip": 0.01131706,
      "auxiliary_loss_mlp": 0.01028274,
      "balance_loss_clip": 1.04826379,
      "balance_loss_mlp": 1.02031064,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 2.0009934497141133,
      "language_loss": 0.72009093,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74169075,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 3.6694447994232178
    },
    {
      "auxiliary_loss_clip": 0.01172843,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.05293512,
      "balance_loss_mlp": 1.02071953,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.18420355329652,
      "language_loss": 0.65254462,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67456043,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.6048314571380615
    },
    {
      "auxiliary_loss_clip": 0.01169018,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.05026257,
      "balance_loss_mlp": 1.01917624,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 1.8149073275959948,
      "language_loss": 0.77264822,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79460919,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.7264556884765625
    },
    {
      "auxiliary_loss_clip": 0.01142399,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.04912639,
      "balance_loss_mlp": 1.0212028,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.890061701464736,
      "language_loss": 0.7864666,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80818522,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.7164249420166016
    },
    {
      "auxiliary_loss_clip": 0.01162927,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.04882252,
      "balance_loss_mlp": 1.02189398,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.8344731726589951,
      "language_loss": 0.75397056,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.7758956,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.7822201251983643
    },
    {
      "auxiliary_loss_clip": 0.01179052,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.05136752,
      "balance_loss_mlp": 1.01948774,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 4.909999086443849,
      "language_loss": 0.6841315,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70620239,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.651203155517578
    },
    {
      "auxiliary_loss_clip": 0.01143048,
      "auxiliary_loss_mlp": 0.00901402,
      "balance_loss_clip": 1.04803085,
      "balance_loss_mlp": 1.00054836,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 1.5323731211600455,
      "language_loss": 0.63820422,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.65864873,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.8446247577667236
    },
    {
      "auxiliary_loss_clip": 0.01064553,
      "auxiliary_loss_mlp": 0.01006245,
      "balance_loss_clip": 1.02080429,
      "balance_loss_mlp": 1.00509477,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6581401863396353,
      "language_loss": 0.52479059,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54549855,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.2750205993652344
    },
    {
      "auxiliary_loss_clip": 0.01166542,
      "auxiliary_loss_mlp": 0.01031683,
      "balance_loss_clip": 1.05273652,
      "balance_loss_mlp": 1.02475095,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 2.5750007458236457,
      "language_loss": 0.72529477,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.74727702,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.734912872314453
    },
    {
      "auxiliary_loss_clip": 0.01064912,
      "auxiliary_loss_mlp": 0.01006168,
      "balance_loss_clip": 1.01453543,
      "balance_loss_mlp": 1.00472534,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.810757648502005,
      "language_loss": 0.5565666,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57727742,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.2216379642486572
    },
    {
      "auxiliary_loss_clip": 0.01178451,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.05185056,
      "balance_loss_mlp": 1.01875889,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.6976194127724613,
      "language_loss": 0.80824649,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.83029544,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.637991189956665
    },
    {
      "auxiliary_loss_clip": 0.0117312,
      "auxiliary_loss_mlp": 0.01029723,
      "balance_loss_clip": 1.05241632,
      "balance_loss_mlp": 1.02192068,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 1.992044990615314,
      "language_loss": 0.88527346,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90730184,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.629746913909912
    },
    {
      "auxiliary_loss_clip": 0.0116,
      "auxiliary_loss_mlp": 0.01028415,
      "balance_loss_clip": 1.04830098,
      "balance_loss_mlp": 1.02029717,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.300431077156822,
      "language_loss": 0.80820888,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.83009303,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.757873058319092
    },
    {
      "auxiliary_loss_clip": 0.01161516,
      "auxiliary_loss_mlp": 0.0090141,
      "balance_loss_clip": 1.04926825,
      "balance_loss_mlp": 1.00058639,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.716332562035707,
      "language_loss": 0.77148449,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.79211378,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 3.580768585205078
    },
    {
      "auxiliary_loss_clip": 0.01164363,
      "auxiliary_loss_mlp": 0.01027694,
      "balance_loss_clip": 1.05007529,
      "balance_loss_mlp": 1.0198617,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 1.9268275057340003,
      "language_loss": 0.7156353,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73755586,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.6284265518188477
    },
    {
      "auxiliary_loss_clip": 0.01179054,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.05246818,
      "balance_loss_mlp": 1.01991129,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.725403481176432,
      "language_loss": 0.79768765,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.81975794,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.6296327114105225
    },
    {
      "auxiliary_loss_clip": 0.01161294,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.04946387,
      "balance_loss_mlp": 1.0235498,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.8630742999318655,
      "language_loss": 0.77462244,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.7965464,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.76401686668396
    },
    {
      "auxiliary_loss_clip": 0.01163703,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.04804361,
      "balance_loss_mlp": 1.02400136,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.7437485650988591,
      "language_loss": 0.78981519,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.81176794,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.751507520675659
    },
    {
      "auxiliary_loss_clip": 0.01142863,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.04550862,
      "balance_loss_mlp": 1.02184188,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.2096012089042696,
      "language_loss": 0.73871613,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76044142,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 3.6510169506073
    },
    {
      "auxiliary_loss_clip": 0.01160997,
      "auxiliary_loss_mlp": 0.01028126,
      "balance_loss_clip": 1.04875231,
      "balance_loss_mlp": 1.02033567,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.6639727192608635,
      "language_loss": 0.8184725,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.84036374,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.74074125289917
    },
    {
      "auxiliary_loss_clip": 0.01153066,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.05139852,
      "balance_loss_mlp": 1.02034545,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 2.285855460648429,
      "language_loss": 0.70053637,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.72235507,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.7762250900268555
    },
    {
      "auxiliary_loss_clip": 0.01185842,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 1.05487561,
      "balance_loss_mlp": 1.01685691,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.419640540236328,
      "language_loss": 0.71576923,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73788249,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 3.542961359024048
    },
    {
      "auxiliary_loss_clip": 0.01169954,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.05173278,
      "balance_loss_mlp": 1.02121782,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 2.296403127592964,
      "language_loss": 0.8189187,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84091175,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.6496105194091797
    },
    {
      "auxiliary_loss_clip": 0.01144384,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.0447855,
      "balance_loss_mlp": 1.01976776,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.4280408698223597,
      "language_loss": 0.73067033,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75239861,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.7056992053985596
    },
    {
      "auxiliary_loss_clip": 0.01169053,
      "auxiliary_loss_mlp": 0.01029512,
      "balance_loss_clip": 1.05146801,
      "balance_loss_mlp": 1.02148366,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.3160206129644245,
      "language_loss": 0.78479564,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80678129,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.6477458477020264
    },
    {
      "auxiliary_loss_clip": 0.01162689,
      "auxiliary_loss_mlp": 0.01027411,
      "balance_loss_clip": 1.0524354,
      "balance_loss_mlp": 1.01969552,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.87639906624726,
      "language_loss": 0.88191658,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90381753,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.6762681007385254
    },
    {
      "auxiliary_loss_clip": 0.01078162,
      "auxiliary_loss_mlp": 0.01010023,
      "balance_loss_clip": 1.01601231,
      "balance_loss_mlp": 1.00863385,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6982459464214686,
      "language_loss": 0.57504332,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59592521,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 4.316842555999756
    },
    {
      "auxiliary_loss_clip": 0.01173825,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.05211353,
      "balance_loss_mlp": 1.01774037,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.5060914436105894,
      "language_loss": 0.83063722,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85263246,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.6904892921447754
    },
    {
      "auxiliary_loss_clip": 0.01177963,
      "auxiliary_loss_mlp": 0.01024343,
      "balance_loss_clip": 1.05046558,
      "balance_loss_mlp": 1.0162847,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 1.997494759209512,
      "language_loss": 0.81105447,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83307743,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.602102041244507
    },
    {
      "auxiliary_loss_clip": 0.0113058,
      "auxiliary_loss_mlp": 0.00901105,
      "balance_loss_clip": 1.04352927,
      "balance_loss_mlp": 1.00054359,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.0822579603310056,
      "language_loss": 0.803092,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.82340884,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.7388713359832764
    },
    {
      "auxiliary_loss_clip": 0.01179313,
      "auxiliary_loss_mlp": 0.01022372,
      "balance_loss_clip": 1.05221748,
      "balance_loss_mlp": 1.01442099,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.393900980550567,
      "language_loss": 0.78328013,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80529702,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.5911500453948975
    },
    {
      "auxiliary_loss_clip": 0.01047695,
      "auxiliary_loss_mlp": 0.01004891,
      "balance_loss_clip": 1.01323986,
      "balance_loss_mlp": 1.00378883,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8452801322685016,
      "language_loss": 0.57322985,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59375572,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.457301616668701
    },
    {
      "auxiliary_loss_clip": 0.01154643,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.04709411,
      "balance_loss_mlp": 1.01861501,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.177894851277846,
      "language_loss": 0.66098058,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68279791,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.7018110752105713
    },
    {
      "auxiliary_loss_clip": 0.01156664,
      "auxiliary_loss_mlp": 0.01031325,
      "balance_loss_clip": 1.04558206,
      "balance_loss_mlp": 1.02252698,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 2.8793858909860424,
      "language_loss": 0.67518795,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.69706786,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.7617194652557373
    },
    {
      "auxiliary_loss_clip": 0.01157152,
      "auxiliary_loss_mlp": 0.01024425,
      "balance_loss_clip": 1.04891872,
      "balance_loss_mlp": 1.01593757,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 1.96210848334262,
      "language_loss": 0.65632254,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67813832,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.717780590057373
    },
    {
      "auxiliary_loss_clip": 0.01166682,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.0482471,
      "balance_loss_mlp": 1.02260911,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.777116776344,
      "language_loss": 0.76403332,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78600359,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.6749508380889893
    },
    {
      "auxiliary_loss_clip": 0.01165707,
      "auxiliary_loss_mlp": 0.01028006,
      "balance_loss_clip": 1.05033708,
      "balance_loss_mlp": 1.01923811,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 2.1760973477852756,
      "language_loss": 0.76209158,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78402871,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.6721508502960205
    },
    {
      "auxiliary_loss_clip": 0.0114843,
      "auxiliary_loss_mlp": 0.00902209,
      "balance_loss_clip": 1.04799414,
      "balance_loss_mlp": 1.00049436,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.863263952889496,
      "language_loss": 0.60695207,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62745845,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.6526968479156494
    },
    {
      "auxiliary_loss_clip": 0.01159157,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.04761755,
      "balance_loss_mlp": 1.02114916,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.136227595483182,
      "language_loss": 0.86021173,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.88210052,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.626288652420044
    },
    {
      "auxiliary_loss_clip": 0.01168697,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.0502944,
      "balance_loss_mlp": 1.0211587,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.6908630443557153,
      "language_loss": 0.79366052,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81564164,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 3.526775598526001
    },
    {
      "auxiliary_loss_clip": 0.01168226,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 1.0506376,
      "balance_loss_mlp": 1.01659393,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 2.2521398128510244,
      "language_loss": 0.83104813,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.8529771,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.653615951538086
    },
    {
      "auxiliary_loss_clip": 0.01168493,
      "auxiliary_loss_mlp": 0.01032667,
      "balance_loss_clip": 1.04883814,
      "balance_loss_mlp": 1.02346992,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.6696643268972964,
      "language_loss": 0.74326122,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.76527286,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.6662960052490234
    },
    {
      "auxiliary_loss_clip": 0.0116053,
      "auxiliary_loss_mlp": 0.01028514,
      "balance_loss_clip": 1.04787731,
      "balance_loss_mlp": 1.02038383,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.5181348920451967,
      "language_loss": 0.80640674,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.82829714,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.7339859008789062
    },
    {
      "auxiliary_loss_clip": 0.01166642,
      "auxiliary_loss_mlp": 0.01039455,
      "balance_loss_clip": 1.05502427,
      "balance_loss_mlp": 1.03136671,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 1.9153892429935917,
      "language_loss": 0.74294579,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76500678,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.6588504314422607
    },
    {
      "auxiliary_loss_clip": 0.01156434,
      "auxiliary_loss_mlp": 0.01030725,
      "balance_loss_clip": 1.05226135,
      "balance_loss_mlp": 1.02219009,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.264017356990045,
      "language_loss": 0.76029396,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.78216559,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.6747031211853027
    },
    {
      "auxiliary_loss_clip": 0.01171708,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.05233538,
      "balance_loss_mlp": 1.01982534,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 2.0887066030943013,
      "language_loss": 0.74226183,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76425469,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 3.537871837615967
    },
    {
      "auxiliary_loss_clip": 0.01174379,
      "auxiliary_loss_mlp": 0.01029854,
      "balance_loss_clip": 1.05304551,
      "balance_loss_mlp": 1.02141976,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 1.9304530942236648,
      "language_loss": 0.71336401,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.7354064,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.6873421669006348
    },
    {
      "auxiliary_loss_clip": 0.01149467,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.05109859,
      "balance_loss_mlp": 1.01893759,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.277382018840574,
      "language_loss": 0.90350819,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92527938,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 3.6034581661224365
    },
    {
      "auxiliary_loss_clip": 0.01138707,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.04681277,
      "balance_loss_mlp": 1.02140713,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.8942198302856763,
      "language_loss": 0.90647244,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92815328,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.726593494415283
    },
    {
      "auxiliary_loss_clip": 0.01150049,
      "auxiliary_loss_mlp": 0.01031193,
      "balance_loss_clip": 1.05121446,
      "balance_loss_mlp": 1.02295542,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 1.9449915453077298,
      "language_loss": 0.80718684,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82899928,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.6995999813079834
    },
    {
      "auxiliary_loss_clip": 0.01153686,
      "auxiliary_loss_mlp": 0.01033614,
      "balance_loss_clip": 1.0543468,
      "balance_loss_mlp": 1.02481985,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 1.826195670104286,
      "language_loss": 0.79290581,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.81477886,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.7109923362731934
    },
    {
      "auxiliary_loss_clip": 0.01172526,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.05250692,
      "balance_loss_mlp": 1.02022398,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 1.9803071030779307,
      "language_loss": 0.80914438,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.83115041,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.598933219909668
    },
    {
      "auxiliary_loss_clip": 0.01174783,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.05057132,
      "balance_loss_mlp": 1.0181179,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.38846883658263,
      "language_loss": 0.69194245,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.7139551,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 3.524087905883789
    },
    {
      "auxiliary_loss_clip": 0.01175484,
      "auxiliary_loss_mlp": 0.01030368,
      "balance_loss_clip": 1.05250311,
      "balance_loss_mlp": 1.02176726,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 2.0125622598555237,
      "language_loss": 0.68766367,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.70972216,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.684891939163208
    },
    {
      "auxiliary_loss_clip": 0.01172639,
      "auxiliary_loss_mlp": 0.01039793,
      "balance_loss_clip": 1.05289578,
      "balance_loss_mlp": 1.03107882,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.0685529263933433,
      "language_loss": 0.7049799,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72710425,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.576636552810669
    },
    {
      "auxiliary_loss_clip": 0.01162304,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 1.0538224,
      "balance_loss_mlp": 1.01866722,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.8832982179837736,
      "language_loss": 0.80604982,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82793921,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.650055408477783
    },
    {
      "auxiliary_loss_clip": 0.01184266,
      "auxiliary_loss_mlp": 0.010272,
      "balance_loss_clip": 1.05294836,
      "balance_loss_mlp": 1.01908779,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 4.746385449646348,
      "language_loss": 0.67872387,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70083857,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.6175649166107178
    },
    {
      "auxiliary_loss_clip": 0.01183527,
      "auxiliary_loss_mlp": 0.01031967,
      "balance_loss_clip": 1.05179346,
      "balance_loss_mlp": 1.02324128,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.4776317549455658,
      "language_loss": 0.66429794,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68645293,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.5804085731506348
    },
    {
      "auxiliary_loss_clip": 0.01170191,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.05068421,
      "balance_loss_mlp": 1.01670659,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.8102255548057884,
      "language_loss": 0.76764989,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.7895987,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.704562187194824
    },
    {
      "auxiliary_loss_clip": 0.01180928,
      "auxiliary_loss_mlp": 0.01028681,
      "balance_loss_clip": 1.05416489,
      "balance_loss_mlp": 1.02080667,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 2.4451183105705314,
      "language_loss": 0.76672208,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.78881818,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.6437296867370605
    },
    {
      "auxiliary_loss_clip": 0.01174145,
      "auxiliary_loss_mlp": 0.00901174,
      "balance_loss_clip": 1.05351007,
      "balance_loss_mlp": 1.00055623,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.44188286165929,
      "language_loss": 0.70765162,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.72840482,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.6025781631469727
    },
    {
      "auxiliary_loss_clip": 0.01162781,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.05102181,
      "balance_loss_mlp": 1.01672459,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.4855485546598355,
      "language_loss": 0.78682989,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80870545,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7640750408172607
    },
    {
      "auxiliary_loss_clip": 0.01155404,
      "auxiliary_loss_mlp": 0.01026064,
      "balance_loss_clip": 1.04703498,
      "balance_loss_mlp": 1.01876211,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 2.078171639004944,
      "language_loss": 0.66765827,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68947291,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.7048943042755127
    },
    {
      "auxiliary_loss_clip": 0.01163832,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.05073965,
      "balance_loss_mlp": 1.01758814,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 2.123504713073778,
      "language_loss": 0.82710528,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84900153,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.6543893814086914
    },
    {
      "auxiliary_loss_clip": 0.01170604,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.05137646,
      "balance_loss_mlp": 1.01630926,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.6744521374222063,
      "language_loss": 0.8382287,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86017066,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.610700845718384
    },
    {
      "auxiliary_loss_clip": 0.01165709,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.05230427,
      "balance_loss_mlp": 1.01945341,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.6955446244500267,
      "language_loss": 0.63499683,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65692198,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 3.6842029094696045
    },
    {
      "auxiliary_loss_clip": 0.01157983,
      "auxiliary_loss_mlp": 0.01023421,
      "balance_loss_clip": 1.04790354,
      "balance_loss_mlp": 1.01600909,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 2.323465712342244,
      "language_loss": 0.70863122,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73044527,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.7212626934051514
    },
    {
      "auxiliary_loss_clip": 0.01141756,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.04267001,
      "balance_loss_mlp": 1.01876533,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.3604434879629546,
      "language_loss": 0.73913985,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76082563,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.676104784011841
    },
    {
      "auxiliary_loss_clip": 0.01183567,
      "auxiliary_loss_mlp": 0.01024721,
      "balance_loss_clip": 1.05332768,
      "balance_loss_mlp": 1.01632833,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 2.1003684248280656,
      "language_loss": 0.76053995,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78262281,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.5844876766204834
    },
    {
      "auxiliary_loss_clip": 0.01159134,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05150998,
      "balance_loss_mlp": 1.02216935,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.86030902094252,
      "language_loss": 0.8082335,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83012116,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.644730567932129
    },
    {
      "auxiliary_loss_clip": 0.01168056,
      "auxiliary_loss_mlp": 0.01025632,
      "balance_loss_clip": 1.04969251,
      "balance_loss_mlp": 1.01751947,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 3.5009004136709376,
      "language_loss": 0.65653664,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67847347,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.6003081798553467
    },
    {
      "auxiliary_loss_clip": 0.01163643,
      "auxiliary_loss_mlp": 0.01021091,
      "balance_loss_clip": 1.05055022,
      "balance_loss_mlp": 1.01375997,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 1.8103279413993911,
      "language_loss": 0.74781597,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.76966339,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 3.551712989807129
    },
    {
      "auxiliary_loss_clip": 0.0117804,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 1.0562582,
      "balance_loss_mlp": 1.01995122,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 5.116698860124139,
      "language_loss": 0.85951686,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88158357,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.657266855239868
    },
    {
      "auxiliary_loss_clip": 0.01172576,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.05104089,
      "balance_loss_mlp": 1.01738536,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 2.2602571097834137,
      "language_loss": 0.71451998,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.7365073,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.5867066383361816
    },
    {
      "auxiliary_loss_clip": 0.01141575,
      "auxiliary_loss_mlp": 0.01028902,
      "balance_loss_clip": 1.04383826,
      "balance_loss_mlp": 1.0202117,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.2407893023764567,
      "language_loss": 0.75336707,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77507186,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 3.88254976272583
    },
    {
      "auxiliary_loss_clip": 0.01144863,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.04700303,
      "balance_loss_mlp": 1.02240002,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.147942055588015,
      "language_loss": 0.77112538,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79288495,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.7871952056884766
    },
    {
      "auxiliary_loss_clip": 0.01179574,
      "auxiliary_loss_mlp": 0.010306,
      "balance_loss_clip": 1.05371022,
      "balance_loss_mlp": 1.0222491,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.7952640796420525,
      "language_loss": 0.48824829,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.51035011,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.614208221435547
    },
    {
      "auxiliary_loss_clip": 0.01174485,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04868293,
      "balance_loss_mlp": 1.01920843,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.1780746020312955,
      "language_loss": 0.78533554,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.8073529,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.617576837539673
    },
    {
      "auxiliary_loss_clip": 0.0116129,
      "auxiliary_loss_mlp": 0.01028888,
      "balance_loss_clip": 1.05239391,
      "balance_loss_mlp": 1.02133346,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.732701059585638,
      "language_loss": 0.75552928,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77743107,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.6124448776245117
    },
    {
      "auxiliary_loss_clip": 0.01070083,
      "auxiliary_loss_mlp": 0.0100144,
      "balance_loss_clip": 1.01778483,
      "balance_loss_mlp": 1.0000155,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.9641494107233424,
      "language_loss": 0.62587333,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64658856,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 4.035198450088501
    },
    {
      "auxiliary_loss_clip": 0.01182285,
      "auxiliary_loss_mlp": 0.01023086,
      "balance_loss_clip": 1.05314529,
      "balance_loss_mlp": 1.01565289,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 2.119442704830068,
      "language_loss": 0.69582796,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71788174,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.5835306644439697
    },
    {
      "auxiliary_loss_clip": 0.01065697,
      "auxiliary_loss_mlp": 0.01002066,
      "balance_loss_clip": 1.01904869,
      "balance_loss_mlp": 1.00074828,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8485078327314188,
      "language_loss": 0.66725492,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68793255,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.1609854698181152
    },
    {
      "auxiliary_loss_clip": 0.0118038,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.05202007,
      "balance_loss_mlp": 1.01655388,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.5583095984947497,
      "language_loss": 0.72908664,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.75113976,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.606426954269409
    },
    {
      "auxiliary_loss_clip": 0.01157135,
      "auxiliary_loss_mlp": 0.01038256,
      "balance_loss_clip": 1.04522622,
      "balance_loss_mlp": 1.02948833,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.452918084292866,
      "language_loss": 0.64765948,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.66961336,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.724635601043701
    },
    {
      "auxiliary_loss_clip": 0.01151654,
      "auxiliary_loss_mlp": 0.01025465,
      "balance_loss_clip": 1.04495263,
      "balance_loss_mlp": 1.01764488,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.5849174433245308,
      "language_loss": 0.85248476,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87425596,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.789929151535034
    },
    {
      "auxiliary_loss_clip": 0.01171983,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 1.04999876,
      "balance_loss_mlp": 1.01839042,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 2.0405920643115527,
      "language_loss": 0.80515558,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82714063,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.679823875427246
    },
    {
      "auxiliary_loss_clip": 0.01165765,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.05203092,
      "balance_loss_mlp": 1.02066803,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.0432989824013954,
      "language_loss": 0.83103579,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.85297477,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.671882152557373
    },
    {
      "auxiliary_loss_clip": 0.01153907,
      "auxiliary_loss_mlp": 0.01022577,
      "balance_loss_clip": 1.04554307,
      "balance_loss_mlp": 1.01401734,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.9152620262084223,
      "language_loss": 0.7310589,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75282371,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.71802020072937
    },
    {
      "auxiliary_loss_clip": 0.01170458,
      "auxiliary_loss_mlp": 0.01031074,
      "balance_loss_clip": 1.05084801,
      "balance_loss_mlp": 1.02262807,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.5992292464908395,
      "language_loss": 0.84268069,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86469603,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.6941168308258057
    },
    {
      "auxiliary_loss_clip": 0.01173021,
      "auxiliary_loss_mlp": 0.01034237,
      "balance_loss_clip": 1.05071425,
      "balance_loss_mlp": 1.02515912,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 3.0221451565570217,
      "language_loss": 0.58835518,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.6104278,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.6649253368377686
    },
    {
      "auxiliary_loss_clip": 0.01155191,
      "auxiliary_loss_mlp": 0.01028419,
      "balance_loss_clip": 1.04905212,
      "balance_loss_mlp": 1.02001476,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 2.614605995047968,
      "language_loss": 0.82983756,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.85167366,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.7007739543914795
    },
    {
      "auxiliary_loss_clip": 0.01179331,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.05545592,
      "balance_loss_mlp": 1.01967311,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 2.1274226992159697,
      "language_loss": 0.84894812,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.8710196,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.6752266883850098
    },
    {
      "auxiliary_loss_clip": 0.01166512,
      "auxiliary_loss_mlp": 0.01029789,
      "balance_loss_clip": 1.05171645,
      "balance_loss_mlp": 1.02205777,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.1096759455992986,
      "language_loss": 0.73376811,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75573111,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 3.5880067348480225
    },
    {
      "auxiliary_loss_clip": 0.01161757,
      "auxiliary_loss_mlp": 0.01023501,
      "balance_loss_clip": 1.05128837,
      "balance_loss_mlp": 1.01591969,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 3.32700304080641,
      "language_loss": 0.67165375,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69350636,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.7653229236602783
    },
    {
      "auxiliary_loss_clip": 0.01170324,
      "auxiliary_loss_mlp": 0.01034467,
      "balance_loss_clip": 1.04871941,
      "balance_loss_mlp": 1.02572894,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.6774405874601257,
      "language_loss": 0.77383375,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79588163,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.5743086338043213
    },
    {
      "auxiliary_loss_clip": 0.01166447,
      "auxiliary_loss_mlp": 0.01031279,
      "balance_loss_clip": 1.05092001,
      "balance_loss_mlp": 1.02174234,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 3.8567128191846822,
      "language_loss": 0.81761771,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.83959496,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.707073926925659
    },
    {
      "auxiliary_loss_clip": 0.01161728,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.04962015,
      "balance_loss_mlp": 1.02174306,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 3.1199139081977587,
      "language_loss": 0.66472656,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68663913,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.758315324783325
    },
    {
      "auxiliary_loss_clip": 0.01145888,
      "auxiliary_loss_mlp": 0.01028295,
      "balance_loss_clip": 1.04654646,
      "balance_loss_mlp": 1.01971173,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.225588406109,
      "language_loss": 0.81426036,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83600217,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 3.6507365703582764
    },
    {
      "auxiliary_loss_clip": 0.011856,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.0579114,
      "balance_loss_mlp": 1.01866651,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 2.2693005928653482,
      "language_loss": 0.8105365,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83265996,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.673706531524658
    },
    {
      "auxiliary_loss_clip": 0.01186572,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.05586028,
      "balance_loss_mlp": 1.0248729,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.9421201796474215,
      "language_loss": 0.790187,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.81238514,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.595906972885132
    },
    {
      "auxiliary_loss_clip": 0.01138303,
      "auxiliary_loss_mlp": 0.01023895,
      "balance_loss_clip": 1.04512739,
      "balance_loss_mlp": 1.01598501,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.9174052821662662,
      "language_loss": 0.76223278,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78385472,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.774301767349243
    },
    {
      "auxiliary_loss_clip": 0.01164424,
      "auxiliary_loss_mlp": 0.01033564,
      "balance_loss_clip": 1.05004716,
      "balance_loss_mlp": 1.02528524,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.9240154796351214,
      "language_loss": 0.76480007,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78678,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 4.000835657119751
    },
    {
      "auxiliary_loss_clip": 0.0117443,
      "auxiliary_loss_mlp": 0.0102384,
      "balance_loss_clip": 1.05463874,
      "balance_loss_mlp": 1.0162884,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.6346862667719857,
      "language_loss": 0.8167882,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83877087,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.6379733085632324
    },
    {
      "auxiliary_loss_clip": 0.01171384,
      "auxiliary_loss_mlp": 0.01023747,
      "balance_loss_clip": 1.05305219,
      "balance_loss_mlp": 1.01613593,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.7572360441455996,
      "language_loss": 0.82215631,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84410763,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.6902873516082764
    },
    {
      "auxiliary_loss_clip": 0.01177791,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.05237269,
      "balance_loss_mlp": 1.01943684,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.058093934817592,
      "language_loss": 0.80609119,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.82814324,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.6766602993011475
    },
    {
      "auxiliary_loss_clip": 0.01147582,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.04789019,
      "balance_loss_mlp": 1.01971292,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.06488620593072,
      "language_loss": 0.69069135,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71244895,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 3.7210495471954346
    },
    {
      "auxiliary_loss_clip": 0.01059447,
      "auxiliary_loss_mlp": 0.01002997,
      "balance_loss_clip": 1.01935518,
      "balance_loss_mlp": 1.00138748,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7444784027160705,
      "language_loss": 0.55483377,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57545829,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.4350922107696533
    },
    {
      "auxiliary_loss_clip": 0.01165189,
      "auxiliary_loss_mlp": 0.0102663,
      "balance_loss_clip": 1.04737484,
      "balance_loss_mlp": 1.01863122,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.7698295998252136,
      "language_loss": 0.71669364,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.73861188,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.781999111175537
    },
    {
      "auxiliary_loss_clip": 0.01159117,
      "auxiliary_loss_mlp": 0.01026171,
      "balance_loss_clip": 1.05075383,
      "balance_loss_mlp": 1.01904476,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.7496876325789208,
      "language_loss": 0.68209559,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70394844,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.7826294898986816
    },
    {
      "auxiliary_loss_clip": 0.01148635,
      "auxiliary_loss_mlp": 0.01029204,
      "balance_loss_clip": 1.04502201,
      "balance_loss_mlp": 1.0200069,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 2.107735611565787,
      "language_loss": 0.81617546,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83795387,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.7475883960723877
    },
    {
      "auxiliary_loss_clip": 0.01166156,
      "auxiliary_loss_mlp": 0.01030821,
      "balance_loss_clip": 1.05541921,
      "balance_loss_mlp": 1.02244639,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.5084782660886895,
      "language_loss": 0.79031682,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81228656,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.7773067951202393
    },
    {
      "auxiliary_loss_clip": 0.01160053,
      "auxiliary_loss_mlp": 0.01026039,
      "balance_loss_clip": 1.05002022,
      "balance_loss_mlp": 1.0179801,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 2.3554201887552333,
      "language_loss": 0.82196742,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84382832,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.7738094329833984
    },
    {
      "auxiliary_loss_clip": 0.01173435,
      "auxiliary_loss_mlp": 0.01027597,
      "balance_loss_clip": 1.05135727,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.8979612187828674,
      "language_loss": 0.74514604,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.7671563,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.6117475032806396
    },
    {
      "auxiliary_loss_clip": 0.01175987,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.05387473,
      "balance_loss_mlp": 1.02265739,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.695441774747941,
      "language_loss": 0.76946694,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79153252,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.676987409591675
    },
    {
      "auxiliary_loss_clip": 0.01154528,
      "auxiliary_loss_mlp": 0.01031232,
      "balance_loss_clip": 1.04546094,
      "balance_loss_mlp": 1.02291107,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 2.101542600145666,
      "language_loss": 0.82841057,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85026819,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.7371864318847656
    },
    {
      "auxiliary_loss_clip": 0.01168591,
      "auxiliary_loss_mlp": 0.01026346,
      "balance_loss_clip": 1.04872251,
      "balance_loss_mlp": 1.01867485,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.295240808809283,
      "language_loss": 0.84391814,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.8658675,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.6532204151153564
    },
    {
      "auxiliary_loss_clip": 0.01066236,
      "auxiliary_loss_mlp": 0.00890886,
      "balance_loss_clip": 1.01922131,
      "balance_loss_mlp": 0.99984854,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.895955036254965,
      "language_loss": 0.55667675,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57624793,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.2045516967773438
    },
    {
      "auxiliary_loss_clip": 0.0116663,
      "auxiliary_loss_mlp": 0.0090124,
      "balance_loss_clip": 1.05441594,
      "balance_loss_mlp": 1.00068581,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.980959684328777,
      "language_loss": 0.83133376,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85201252,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.6802282333374023
    },
    {
      "auxiliary_loss_clip": 0.01169879,
      "auxiliary_loss_mlp": 0.00901593,
      "balance_loss_clip": 1.05162871,
      "balance_loss_mlp": 1.00070524,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.7553190728924801,
      "language_loss": 0.76828825,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.7890029,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.574652910232544
    },
    {
      "auxiliary_loss_clip": 0.01149535,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 1.04687631,
      "balance_loss_mlp": 1.01869941,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.9830040903849606,
      "language_loss": 0.69676232,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.71853411,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 4.018509864807129
    },
    {
      "auxiliary_loss_clip": 0.01178992,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.05279374,
      "balance_loss_mlp": 1.02247334,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.5313895938230146,
      "language_loss": 0.89054066,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91263652,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.704909086227417
    },
    {
      "auxiliary_loss_clip": 0.01176428,
      "auxiliary_loss_mlp": 0.01038462,
      "balance_loss_clip": 1.0545013,
      "balance_loss_mlp": 1.02943802,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.7451529182385854,
      "language_loss": 0.80478263,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82693154,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.5966267585754395
    },
    {
      "auxiliary_loss_clip": 0.01161227,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.05253851,
      "balance_loss_mlp": 1.01844525,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 2.1985425203040685,
      "language_loss": 0.80087471,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82275128,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.710749864578247
    },
    {
      "auxiliary_loss_clip": 0.01152657,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04884744,
      "balance_loss_mlp": 1.02134871,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.7779969399620774,
      "language_loss": 0.7136184,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73543829,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.707749366760254
    },
    {
      "auxiliary_loss_clip": 0.01164298,
      "auxiliary_loss_mlp": 0.01026216,
      "balance_loss_clip": 1.05411863,
      "balance_loss_mlp": 1.0190661,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 1.7039269481991914,
      "language_loss": 0.84773231,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.86963749,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 3.6039645671844482
    },
    {
      "auxiliary_loss_clip": 0.01157957,
      "auxiliary_loss_mlp": 0.01028988,
      "balance_loss_clip": 1.0502249,
      "balance_loss_mlp": 1.02052999,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 2.076714033341033,
      "language_loss": 0.8350991,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85696852,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.6649045944213867
    },
    {
      "auxiliary_loss_clip": 0.01161016,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.04860044,
      "balance_loss_mlp": 1.01662266,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.098956078914574,
      "language_loss": 0.87327909,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89513934,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.710693359375
    },
    {
      "auxiliary_loss_clip": 0.01170516,
      "auxiliary_loss_mlp": 0.0103018,
      "balance_loss_clip": 1.05492663,
      "balance_loss_mlp": 1.02215135,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 2.1918712415515014,
      "language_loss": 0.84561694,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86762387,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 3.6124773025512695
    },
    {
      "auxiliary_loss_clip": 0.011456,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.04364753,
      "balance_loss_mlp": 1.01747012,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 6.870623408698229,
      "language_loss": 0.76027703,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78198576,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.7199647426605225
    },
    {
      "auxiliary_loss_clip": 0.01156815,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.05017304,
      "balance_loss_mlp": 1.01823139,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.587463921013547,
      "language_loss": 0.74885058,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.77068424,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.782872438430786
    },
    {
      "auxiliary_loss_clip": 0.01171633,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.05077052,
      "balance_loss_mlp": 1.02071452,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 1.9225228726545844,
      "language_loss": 0.83447319,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85647684,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.623859167098999
    },
    {
      "auxiliary_loss_clip": 0.01155557,
      "auxiliary_loss_mlp": 0.00901705,
      "balance_loss_clip": 1.04710042,
      "balance_loss_mlp": 1.0006187,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.9391993577399138,
      "language_loss": 0.72079027,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.74136287,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.73724627494812
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01033293,
      "balance_loss_clip": 1.05605567,
      "balance_loss_mlp": 1.02460313,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.11585190254171,
      "language_loss": 0.73102582,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75303131,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 3.6436383724212646
    },
    {
      "auxiliary_loss_clip": 0.01168683,
      "auxiliary_loss_mlp": 0.00901176,
      "balance_loss_clip": 1.05063689,
      "balance_loss_mlp": 1.00074124,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.7353609337682756,
      "language_loss": 0.74893367,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.76963228,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.7162845134735107
    },
    {
      "auxiliary_loss_clip": 0.01168968,
      "auxiliary_loss_mlp": 0.01032723,
      "balance_loss_clip": 1.05259252,
      "balance_loss_mlp": 1.02397346,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.785773436447859,
      "language_loss": 0.8498466,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.87186348,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.7443535327911377
    },
    {
      "auxiliary_loss_clip": 0.01165827,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.04919064,
      "balance_loss_mlp": 1.02442193,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 1.8739423045957015,
      "language_loss": 0.78068328,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.80266762,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.709343671798706
    },
    {
      "auxiliary_loss_clip": 0.011599,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.0470798,
      "balance_loss_mlp": 1.02035928,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 1.6928146931244816,
      "language_loss": 0.69826198,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.72014678,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.731412410736084
    },
    {
      "auxiliary_loss_clip": 0.01180908,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.05266786,
      "balance_loss_mlp": 1.0178076,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.7598566509356883,
      "language_loss": 0.73724437,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.7593087,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.6253252029418945
    },
    {
      "auxiliary_loss_clip": 0.0116864,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 1.04858577,
      "balance_loss_mlp": 1.0198772,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 2.0090398454678513,
      "language_loss": 0.63403362,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.65600324,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.7145612239837646
    },
    {
      "auxiliary_loss_clip": 0.01187198,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.05667353,
      "balance_loss_mlp": 1.020082,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.8212868903195623,
      "language_loss": 0.78953075,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81168771,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.645186424255371
    },
    {
      "auxiliary_loss_clip": 0.01180146,
      "auxiliary_loss_mlp": 0.00902248,
      "balance_loss_clip": 1.05434287,
      "balance_loss_mlp": 1.00053763,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.175880230334109,
      "language_loss": 0.75942576,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78024971,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.7424049377441406
    },
    {
      "auxiliary_loss_clip": 0.011613,
      "auxiliary_loss_mlp": 0.01025241,
      "balance_loss_clip": 1.05033767,
      "balance_loss_mlp": 1.01779306,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.5664621691903804,
      "language_loss": 0.70406717,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.7259326,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.7596917152404785
    },
    {
      "auxiliary_loss_clip": 0.01188959,
      "auxiliary_loss_mlp": 0.0103495,
      "balance_loss_clip": 1.05514932,
      "balance_loss_mlp": 1.02593231,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 6.44579067217014,
      "language_loss": 0.77968454,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80192363,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.647350311279297
    },
    {
      "auxiliary_loss_clip": 0.01165653,
      "auxiliary_loss_mlp": 0.01028548,
      "balance_loss_clip": 1.05338573,
      "balance_loss_mlp": 1.02046597,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.680419434147726,
      "language_loss": 0.71479136,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73673332,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.7679293155670166
    },
    {
      "auxiliary_loss_clip": 0.01183947,
      "auxiliary_loss_mlp": 0.01030917,
      "balance_loss_clip": 1.05393648,
      "balance_loss_mlp": 1.02251256,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 2.1250447261696888,
      "language_loss": 0.78165746,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80380613,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.6104159355163574
    },
    {
      "auxiliary_loss_clip": 0.0118444,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 1.0554204,
      "balance_loss_mlp": 1.01842022,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.926396724203093,
      "language_loss": 0.70916903,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73127472,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 3.6614890098571777
    },
    {
      "auxiliary_loss_clip": 0.0117329,
      "auxiliary_loss_mlp": 0.01026898,
      "balance_loss_clip": 1.05550551,
      "balance_loss_mlp": 1.01882136,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.8800754834768307,
      "language_loss": 0.72788692,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74988878,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.645176649093628
    },
    {
      "auxiliary_loss_clip": 0.01172166,
      "auxiliary_loss_mlp": 0.01028892,
      "balance_loss_clip": 1.05439472,
      "balance_loss_mlp": 1.02112508,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.925033298596914,
      "language_loss": 0.88298035,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90499091,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.73060941696167
    },
    {
      "auxiliary_loss_clip": 0.01167852,
      "auxiliary_loss_mlp": 0.01032468,
      "balance_loss_clip": 1.05367959,
      "balance_loss_mlp": 1.02312255,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 2.473201542604303,
      "language_loss": 0.69096071,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71296394,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.7442142963409424
    },
    {
      "auxiliary_loss_clip": 0.01175771,
      "auxiliary_loss_mlp": 0.00901709,
      "balance_loss_clip": 1.05190659,
      "balance_loss_mlp": 1.00059593,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 2.1181176236685593,
      "language_loss": 0.80114448,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.82191926,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.6619670391082764
    },
    {
      "auxiliary_loss_clip": 0.01170738,
      "auxiliary_loss_mlp": 0.00901375,
      "balance_loss_clip": 1.05402946,
      "balance_loss_mlp": 1.00053561,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.6691696937588922,
      "language_loss": 0.72298801,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74370909,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 3.54862380027771
    },
    {
      "auxiliary_loss_clip": 0.01173746,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.05605984,
      "balance_loss_mlp": 1.01659751,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 3.1765860988536034,
      "language_loss": 0.83246028,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85444671,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.6339871883392334
    },
    {
      "auxiliary_loss_clip": 0.01139399,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.045035,
      "balance_loss_mlp": 1.02130413,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 2.062618837429423,
      "language_loss": 0.78449649,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.80619097,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.806544542312622
    },
    {
      "auxiliary_loss_clip": 0.01163389,
      "auxiliary_loss_mlp": 0.01029139,
      "balance_loss_clip": 1.05157816,
      "balance_loss_mlp": 1.02022243,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.1844116951345653,
      "language_loss": 0.87771857,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89964384,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.6381778717041016
    },
    {
      "auxiliary_loss_clip": 0.01174972,
      "auxiliary_loss_mlp": 0.01021875,
      "balance_loss_clip": 1.05240667,
      "balance_loss_mlp": 1.01448989,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.5041434431614853,
      "language_loss": 0.76078641,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.7827549,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 3.5911028385162354
    },
    {
      "auxiliary_loss_clip": 0.01154795,
      "auxiliary_loss_mlp": 0.01028399,
      "balance_loss_clip": 1.0489769,
      "balance_loss_mlp": 1.01997697,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 2.2613251369266627,
      "language_loss": 0.68565166,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.70748365,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.7873528003692627
    },
    {
      "auxiliary_loss_clip": 0.01145341,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.0440526,
      "balance_loss_mlp": 1.01854575,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 2.2239862692887047,
      "language_loss": 0.79955208,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82127452,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.734961986541748
    },
    {
      "auxiliary_loss_clip": 0.01164095,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.04694176,
      "balance_loss_mlp": 1.02162468,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.7719709800845536,
      "language_loss": 0.66439843,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.68634009,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.894435405731201
    },
    {
      "auxiliary_loss_clip": 0.01159599,
      "auxiliary_loss_mlp": 0.01033335,
      "balance_loss_clip": 1.0506897,
      "balance_loss_mlp": 1.02552676,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.3257793021723003,
      "language_loss": 0.85709375,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87902313,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 3.7098071575164795
    },
    {
      "auxiliary_loss_clip": 0.01150024,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.04661226,
      "balance_loss_mlp": 1.0238862,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.9175659777120257,
      "language_loss": 0.89784271,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91966462,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.726898670196533
    },
    {
      "auxiliary_loss_clip": 0.01166481,
      "auxiliary_loss_mlp": 0.00901496,
      "balance_loss_clip": 1.05228007,
      "balance_loss_mlp": 1.00062621,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.9407895587704798,
      "language_loss": 0.75122023,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.7719,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.734180450439453
    },
    {
      "auxiliary_loss_clip": 0.01170831,
      "auxiliary_loss_mlp": 0.01029397,
      "balance_loss_clip": 1.05513859,
      "balance_loss_mlp": 1.02059305,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 2.1035209658825575,
      "language_loss": 0.75743103,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77943337,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.6407530307769775
    },
    {
      "auxiliary_loss_clip": 0.01138123,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.04911649,
      "balance_loss_mlp": 1.01905835,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.5494506428416133,
      "language_loss": 0.77134538,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79300129,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.81907320022583
    },
    {
      "auxiliary_loss_clip": 0.0118019,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.05452394,
      "balance_loss_mlp": 1.02094889,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.6757091058205524,
      "language_loss": 0.77181673,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79390466,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.624682903289795
    },
    {
      "auxiliary_loss_clip": 0.01171621,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.05105782,
      "balance_loss_mlp": 1.02172971,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.794881038587774,
      "language_loss": 0.76802444,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.79003644,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.6850080490112305
    },
    {
      "auxiliary_loss_clip": 0.0118209,
      "auxiliary_loss_mlp": 0.01031044,
      "balance_loss_clip": 1.0516994,
      "balance_loss_mlp": 1.02236521,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.9045761880568408,
      "language_loss": 0.80985075,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.83198214,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.5501723289489746
    },
    {
      "auxiliary_loss_clip": 0.01183259,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.05397129,
      "balance_loss_mlp": 1.01954389,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 4.616984065802376,
      "language_loss": 0.66284633,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68495548,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.612607479095459
    },
    {
      "auxiliary_loss_clip": 0.01155268,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04964924,
      "balance_loss_mlp": 1.01681554,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 1.8032368711574034,
      "language_loss": 0.80750716,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82931173,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.6123569011688232
    },
    {
      "auxiliary_loss_clip": 0.01173878,
      "auxiliary_loss_mlp": 0.01034446,
      "balance_loss_clip": 1.05108154,
      "balance_loss_mlp": 1.02620912,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 3.5672227689822202,
      "language_loss": 0.78771019,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80979347,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.706014394760132
    },
    {
      "auxiliary_loss_clip": 0.01149426,
      "auxiliary_loss_mlp": 0.01033728,
      "balance_loss_clip": 1.04671347,
      "balance_loss_mlp": 1.02520466,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 2.052016531750145,
      "language_loss": 0.83487695,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85670853,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.714322090148926
    },
    {
      "auxiliary_loss_clip": 0.01183209,
      "auxiliary_loss_mlp": 0.01026456,
      "balance_loss_clip": 1.05468798,
      "balance_loss_mlp": 1.01896977,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 1.8058462620535147,
      "language_loss": 0.81785208,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.83994865,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.6202521324157715
    },
    {
      "auxiliary_loss_clip": 0.01101497,
      "auxiliary_loss_mlp": 0.01029133,
      "balance_loss_clip": 1.04095769,
      "balance_loss_mlp": 1.02205157,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.138564001647884,
      "language_loss": 0.83141959,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.85272586,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 3.8268866539001465
    },
    {
      "auxiliary_loss_clip": 0.01130492,
      "auxiliary_loss_mlp": 0.01034102,
      "balance_loss_clip": 1.04795647,
      "balance_loss_mlp": 1.02596641,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.863585861269734,
      "language_loss": 0.78232777,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80397373,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.777453660964966
    },
    {
      "auxiliary_loss_clip": 0.01189192,
      "auxiliary_loss_mlp": 0.01031657,
      "balance_loss_clip": 1.05705261,
      "balance_loss_mlp": 1.02298462,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 4.005426449515995,
      "language_loss": 0.86832893,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.89053738,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.708503484725952
    },
    {
      "auxiliary_loss_clip": 0.01179856,
      "auxiliary_loss_mlp": 0.00901046,
      "balance_loss_clip": 1.05691779,
      "balance_loss_mlp": 1.00048256,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 1.8153747932265798,
      "language_loss": 0.80119848,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.82200754,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.5890862941741943
    },
    {
      "auxiliary_loss_clip": 0.01169405,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.0487206,
      "balance_loss_mlp": 1.01941979,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.2344126502592845,
      "language_loss": 0.73434937,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75632107,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.6391401290893555
    },
    {
      "auxiliary_loss_clip": 0.01168585,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.05374074,
      "balance_loss_mlp": 1.01829672,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.8868883322745635,
      "language_loss": 0.77372986,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79568696,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.804586172103882
    },
    {
      "auxiliary_loss_clip": 0.01185062,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.05707455,
      "balance_loss_mlp": 1.02249372,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.0604644875597464,
      "language_loss": 0.82465744,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84680772,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 3.4832499027252197
    },
    {
      "auxiliary_loss_clip": 0.01160178,
      "auxiliary_loss_mlp": 0.00901571,
      "balance_loss_clip": 1.04995775,
      "balance_loss_mlp": 1.00061011,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.235820681578995,
      "language_loss": 0.78132707,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80194455,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.7260971069335938
    },
    {
      "auxiliary_loss_clip": 0.01185384,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.05472255,
      "balance_loss_mlp": 1.02157044,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 1.8363451178355752,
      "language_loss": 0.85402852,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87618315,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.5766007900238037
    },
    {
      "auxiliary_loss_clip": 0.01170775,
      "auxiliary_loss_mlp": 0.01028262,
      "balance_loss_clip": 1.05341518,
      "balance_loss_mlp": 1.01981604,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.218055882645009,
      "language_loss": 0.82276565,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.84475601,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 3.5542094707489014
    },
    {
      "auxiliary_loss_clip": 0.01085608,
      "auxiliary_loss_mlp": 0.01004692,
      "balance_loss_clip": 1.01957595,
      "balance_loss_mlp": 1.00345838,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.8973846875623243,
      "language_loss": 0.63171673,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65261972,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.288200855255127
    },
    {
      "auxiliary_loss_clip": 0.01165903,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05135369,
      "balance_loss_mlp": 1.02330732,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.6541139704626544,
      "language_loss": 0.76752657,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.78950059,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.68241024017334
    },
    {
      "auxiliary_loss_clip": 0.01165999,
      "auxiliary_loss_mlp": 0.01031258,
      "balance_loss_clip": 1.04916978,
      "balance_loss_mlp": 1.0223825,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.796056382572249,
      "language_loss": 0.76172578,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.78369832,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.672698736190796
    },
    {
      "auxiliary_loss_clip": 0.01167142,
      "auxiliary_loss_mlp": 0.01034689,
      "balance_loss_clip": 1.05292082,
      "balance_loss_mlp": 1.02575445,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.9055105299370552,
      "language_loss": 0.84844035,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.87045866,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.6538147926330566
    },
    {
      "auxiliary_loss_clip": 0.01168812,
      "auxiliary_loss_mlp": 0.01028276,
      "balance_loss_clip": 1.05359709,
      "balance_loss_mlp": 1.0197463,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.6564752881617548,
      "language_loss": 0.79111719,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81308806,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 3.612586498260498
    },
    {
      "auxiliary_loss_clip": 0.01154049,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04941809,
      "balance_loss_mlp": 1.02073884,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.7893880310760129,
      "language_loss": 0.84480566,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86662614,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.6731011867523193
    },
    {
      "auxiliary_loss_clip": 0.01085387,
      "auxiliary_loss_mlp": 0.01001206,
      "balance_loss_clip": 1.01935875,
      "balance_loss_mlp": 1.00005519,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7696577426845822,
      "language_loss": 0.57731467,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59818065,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.0797207355499268
    },
    {
      "auxiliary_loss_clip": 0.01180989,
      "auxiliary_loss_mlp": 0.0102816,
      "balance_loss_clip": 1.05403471,
      "balance_loss_mlp": 1.01992834,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.9990792909998136,
      "language_loss": 0.72895712,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75104856,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.6048266887664795
    },
    {
      "auxiliary_loss_clip": 0.0117308,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.05361986,
      "balance_loss_mlp": 1.02408218,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.5798660562094964,
      "language_loss": 0.72300625,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74506509,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.620631456375122
    },
    {
      "auxiliary_loss_clip": 0.01187527,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 1.05621147,
      "balance_loss_mlp": 1.01760316,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.9799738242866876,
      "language_loss": 0.78006172,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80219811,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.6063852310180664
    },
    {
      "auxiliary_loss_clip": 0.01178438,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.05694854,
      "balance_loss_mlp": 1.01947713,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 4.760732779045024,
      "language_loss": 0.74469578,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76676363,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.6244142055511475
    },
    {
      "auxiliary_loss_clip": 0.01185014,
      "auxiliary_loss_mlp": 0.01026748,
      "balance_loss_clip": 1.05630851,
      "balance_loss_mlp": 1.01878238,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.7574283391475263,
      "language_loss": 0.79169214,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81380975,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.5956382751464844
    },
    {
      "auxiliary_loss_clip": 0.01187523,
      "auxiliary_loss_mlp": 0.01034627,
      "balance_loss_clip": 1.05646801,
      "balance_loss_mlp": 1.02616894,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 4.782424575594474,
      "language_loss": 0.73107862,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.75330013,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.6345114707946777
    },
    {
      "auxiliary_loss_clip": 0.01176227,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.05512309,
      "balance_loss_mlp": 1.02259338,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 2.1504975541087994,
      "language_loss": 0.88136417,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90343213,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.6562161445617676
    },
    {
      "auxiliary_loss_clip": 0.01185651,
      "auxiliary_loss_mlp": 0.01024152,
      "balance_loss_clip": 1.05475879,
      "balance_loss_mlp": 1.0161593,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 3.7326241259940827,
      "language_loss": 0.70645458,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72855258,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.6012535095214844
    },
    {
      "auxiliary_loss_clip": 0.0114145,
      "auxiliary_loss_mlp": 0.01028048,
      "balance_loss_clip": 1.04360533,
      "balance_loss_mlp": 1.01968527,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.6836771334679186,
      "language_loss": 0.73805356,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75974858,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.742553234100342
    },
    {
      "auxiliary_loss_clip": 0.01169937,
      "auxiliary_loss_mlp": 0.00902032,
      "balance_loss_clip": 1.05345178,
      "balance_loss_mlp": 1.00067401,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 2.2940191669876695,
      "language_loss": 0.79120672,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.81192636,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.6328182220458984
    },
    {
      "auxiliary_loss_clip": 0.0117653,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.05283737,
      "balance_loss_mlp": 1.0194447,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.6302770860852192,
      "language_loss": 0.78976107,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81180197,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 3.6570160388946533
    },
    {
      "auxiliary_loss_clip": 0.01176379,
      "auxiliary_loss_mlp": 0.00902273,
      "balance_loss_clip": 1.05442715,
      "balance_loss_mlp": 1.00075519,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 2.0558851626959678,
      "language_loss": 0.75045073,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.77123725,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.695401191711426
    },
    {
      "auxiliary_loss_clip": 0.01177574,
      "auxiliary_loss_mlp": 0.0090128,
      "balance_loss_clip": 1.05705047,
      "balance_loss_mlp": 1.00066555,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.532354294828751,
      "language_loss": 0.76457906,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.78536755,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.943437099456787
    },
    {
      "auxiliary_loss_clip": 0.0114865,
      "auxiliary_loss_mlp": 0.01026686,
      "balance_loss_clip": 1.04657936,
      "balance_loss_mlp": 1.01873469,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 1.8369208063311162,
      "language_loss": 0.75445712,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77621049,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.789767265319824
    },
    {
      "auxiliary_loss_clip": 0.01159624,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.04881263,
      "balance_loss_mlp": 1.02189541,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 3.200094252065562,
      "language_loss": 0.78647649,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.8083784,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.6952731609344482
    },
    {
      "auxiliary_loss_clip": 0.0117657,
      "auxiliary_loss_mlp": 0.01023702,
      "balance_loss_clip": 1.05206656,
      "balance_loss_mlp": 1.01544642,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.532134775866031,
      "language_loss": 0.77444661,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.7964493,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.619745969772339
    },
    {
      "auxiliary_loss_clip": 0.01153135,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.04800832,
      "balance_loss_mlp": 1.0198791,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 1.8026552749448657,
      "language_loss": 0.68339676,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.70520818,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 3.5605411529541016
    },
    {
      "auxiliary_loss_clip": 0.01123242,
      "auxiliary_loss_mlp": 0.00901446,
      "balance_loss_clip": 1.04470015,
      "balance_loss_mlp": 1.00057983,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 1.986955200651342,
      "language_loss": 0.76938969,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78963661,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.7379610538482666
    },
    {
      "auxiliary_loss_clip": 0.01190849,
      "auxiliary_loss_mlp": 0.01037424,
      "balance_loss_clip": 1.05832577,
      "balance_loss_mlp": 1.02904367,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.2510534758084653,
      "language_loss": 0.84454083,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86682355,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.60557222366333
    },
    {
      "auxiliary_loss_clip": 0.01169626,
      "auxiliary_loss_mlp": 0.01025498,
      "balance_loss_clip": 1.05384254,
      "balance_loss_mlp": 1.0171299,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 2.4163135922677847,
      "language_loss": 0.77685535,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79880655,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 3.507498025894165
    },
    {
      "auxiliary_loss_clip": 0.01158887,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.04958773,
      "balance_loss_mlp": 1.0183332,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.1923104562698024,
      "language_loss": 0.6422168,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66407925,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.6812024116516113
    },
    {
      "auxiliary_loss_clip": 0.01170146,
      "auxiliary_loss_mlp": 0.01034476,
      "balance_loss_clip": 1.05351067,
      "balance_loss_mlp": 1.0262686,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.7903326010006217,
      "language_loss": 0.75216186,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77420807,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.650489568710327
    },
    {
      "auxiliary_loss_clip": 0.0118502,
      "auxiliary_loss_mlp": 0.01026239,
      "balance_loss_clip": 1.0549221,
      "balance_loss_mlp": 1.01808536,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 2.11259644493533,
      "language_loss": 0.80593944,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82805192,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.668626070022583
    },
    {
      "auxiliary_loss_clip": 0.01157485,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.04849505,
      "balance_loss_mlp": 1.02054417,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.238248032909881,
      "language_loss": 0.81446767,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83633018,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.6606335639953613
    },
    {
      "auxiliary_loss_clip": 0.01166601,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.05272233,
      "balance_loss_mlp": 1.02220809,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.7865987214236525,
      "language_loss": 0.78648663,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.80845749,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 3.518035411834717
    },
    {
      "auxiliary_loss_clip": 0.01158283,
      "auxiliary_loss_mlp": 0.01027237,
      "balance_loss_clip": 1.05104935,
      "balance_loss_mlp": 1.0187192,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.6894015777862927,
      "language_loss": 0.7415334,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76338863,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.6992878913879395
    },
    {
      "auxiliary_loss_clip": 0.01069082,
      "auxiliary_loss_mlp": 0.01002868,
      "balance_loss_clip": 1.0342921,
      "balance_loss_mlp": 1.00175321,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7413824542216655,
      "language_loss": 0.61579597,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.6365155,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.1837823390960693
    },
    {
      "auxiliary_loss_clip": 0.01142453,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.04522991,
      "balance_loss_mlp": 1.020859,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 1.9983017416353475,
      "language_loss": 0.6092639,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63098335,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.737551212310791
    },
    {
      "auxiliary_loss_clip": 0.01183131,
      "auxiliary_loss_mlp": 0.01030935,
      "balance_loss_clip": 1.05299234,
      "balance_loss_mlp": 1.02290583,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.576621393007635,
      "language_loss": 0.81919503,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.84133571,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.6043906211853027
    },
    {
      "auxiliary_loss_clip": 0.01176598,
      "auxiliary_loss_mlp": 0.0102171,
      "balance_loss_clip": 1.05369699,
      "balance_loss_mlp": 1.01393735,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.750203003830038,
      "language_loss": 0.80821949,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.83020258,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.640162229537964
    },
    {
      "auxiliary_loss_clip": 0.01171083,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.0548991,
      "balance_loss_mlp": 1.02148938,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.9009221779835441,
      "language_loss": 0.8467772,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86878115,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.6643800735473633
    },
    {
      "auxiliary_loss_clip": 0.01172114,
      "auxiliary_loss_mlp": 0.01023409,
      "balance_loss_clip": 1.04978132,
      "balance_loss_mlp": 1.01499248,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.2537671127290326,
      "language_loss": 0.69738317,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.71933842,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.6321094036102295
    },
    {
      "auxiliary_loss_clip": 0.01155359,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.05619311,
      "balance_loss_mlp": 1.02374804,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.6126913461872543,
      "language_loss": 0.74454486,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76641268,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.6401970386505127
    },
    {
      "auxiliary_loss_clip": 0.01183464,
      "auxiliary_loss_mlp": 0.01027051,
      "balance_loss_clip": 1.05367732,
      "balance_loss_mlp": 1.01852119,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.7009747509628916,
      "language_loss": 0.70525014,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.7273553,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.6420559883117676
    },
    {
      "auxiliary_loss_clip": 0.01159016,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.04918158,
      "balance_loss_mlp": 1.02056348,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 2.1029571077515956,
      "language_loss": 0.8492735,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.87115651,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 2.693086862564087
    },
    {
      "auxiliary_loss_clip": 0.01166704,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.04844403,
      "balance_loss_mlp": 1.01767349,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.7050573756044143,
      "language_loss": 0.71172082,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73364663,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.6682381629943848
    },
    {
      "auxiliary_loss_clip": 0.01172,
      "auxiliary_loss_mlp": 0.01025874,
      "balance_loss_clip": 1.05467141,
      "balance_loss_mlp": 1.01808977,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 2.1500943560554897,
      "language_loss": 0.74033189,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.76231062,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 3.5800509452819824
    },
    {
      "auxiliary_loss_clip": 0.01064003,
      "auxiliary_loss_mlp": 0.010035,
      "balance_loss_clip": 1.01685739,
      "balance_loss_mlp": 1.00237298,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.6985128988485952,
      "language_loss": 0.5231235,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54379851,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.169957399368286
    },
    {
      "auxiliary_loss_clip": 0.0106722,
      "auxiliary_loss_mlp": 0.01006233,
      "balance_loss_clip": 1.01841331,
      "balance_loss_mlp": 1.00501716,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7180000655005129,
      "language_loss": 0.54427528,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56500983,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.3168606758117676
    },
    {
      "auxiliary_loss_clip": 0.01177211,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.05361903,
      "balance_loss_mlp": 1.01938736,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.9216461828156852,
      "language_loss": 0.75646383,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77850616,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.6427841186523438
    },
    {
      "auxiliary_loss_clip": 0.01162759,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04936981,
      "balance_loss_mlp": 1.01931381,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 3.4611984366522983,
      "language_loss": 0.84451157,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86641467,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.70858097076416
    },
    {
      "auxiliary_loss_clip": 0.01175608,
      "auxiliary_loss_mlp": 0.01026202,
      "balance_loss_clip": 1.05480242,
      "balance_loss_mlp": 1.017977,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 1.9617287285344456,
      "language_loss": 0.82921273,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85123086,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.6534736156463623
    },
    {
      "auxiliary_loss_clip": 0.01153816,
      "auxiliary_loss_mlp": 0.01023703,
      "balance_loss_clip": 1.04918659,
      "balance_loss_mlp": 1.01565671,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 3.520910457785773,
      "language_loss": 0.83192444,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85369962,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.673102855682373
    },
    {
      "auxiliary_loss_clip": 0.0116608,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.05578661,
      "balance_loss_mlp": 1.01976752,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.4657586737220534,
      "language_loss": 0.77097738,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79291749,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 3.551926612854004
    },
    {
      "auxiliary_loss_clip": 0.01164786,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.05166054,
      "balance_loss_mlp": 1.01881981,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 1.905164371812367,
      "language_loss": 0.70718938,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.72911,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.7261743545532227
    },
    {
      "auxiliary_loss_clip": 0.01165732,
      "auxiliary_loss_mlp": 0.01026145,
      "balance_loss_clip": 1.05055308,
      "balance_loss_mlp": 1.017097,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.8338641417297645,
      "language_loss": 0.78175813,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.8036769,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 3.608574151992798
    },
    {
      "auxiliary_loss_clip": 0.01173775,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.05329645,
      "balance_loss_mlp": 1.01972055,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 1.9829267351008923,
      "language_loss": 0.77324587,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79525638,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.6113884449005127
    },
    {
      "auxiliary_loss_clip": 0.01178372,
      "auxiliary_loss_mlp": 0.01030248,
      "balance_loss_clip": 1.05588031,
      "balance_loss_mlp": 1.02224267,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.9046033047797,
      "language_loss": 0.85286951,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87495571,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.6470038890838623
    },
    {
      "auxiliary_loss_clip": 0.01161067,
      "auxiliary_loss_mlp": 0.01027259,
      "balance_loss_clip": 1.05181456,
      "balance_loss_mlp": 1.01894403,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 1.864005607350872,
      "language_loss": 0.72486663,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74674988,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.657925844192505
    },
    {
      "auxiliary_loss_clip": 0.01160879,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.0517199,
      "balance_loss_mlp": 1.02000737,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 2.508522649325554,
      "language_loss": 0.84520066,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86709684,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.7793145179748535
    },
    {
      "auxiliary_loss_clip": 0.01175507,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.05086529,
      "balance_loss_mlp": 1.01562524,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 2.090296885824719,
      "language_loss": 0.77622032,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.79821551,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 3.593538522720337
    },
    {
      "auxiliary_loss_clip": 0.01174966,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.05363727,
      "balance_loss_mlp": 1.01646614,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.216676562558015,
      "language_loss": 0.76100087,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78299916,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.6097700595855713
    },
    {
      "auxiliary_loss_clip": 0.01059875,
      "auxiliary_loss_mlp": 0.01002947,
      "balance_loss_clip": 1.01642382,
      "balance_loss_mlp": 1.00187445,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8330340259455499,
      "language_loss": 0.57886171,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.59948993,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.2997970581054688
    },
    {
      "auxiliary_loss_clip": 0.01179278,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.05252218,
      "balance_loss_mlp": 1.02245796,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 2.080180459891597,
      "language_loss": 0.79287505,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81497204,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.650022506713867
    },
    {
      "auxiliary_loss_clip": 0.01175502,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.05174553,
      "balance_loss_mlp": 1.01859915,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 3.1152459096559606,
      "language_loss": 0.74181521,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76383907,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.6257264614105225
    },
    {
      "auxiliary_loss_clip": 0.01176555,
      "auxiliary_loss_mlp": 0.01028164,
      "balance_loss_clip": 1.05273294,
      "balance_loss_mlp": 1.02042186,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 1.884451602655554,
      "language_loss": 0.75102979,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77307701,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.7160608768463135
    },
    {
      "auxiliary_loss_clip": 0.01146952,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.04979968,
      "balance_loss_mlp": 1.01618695,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 2.6201328136551374,
      "language_loss": 0.69367796,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71539265,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.701206922531128
    },
    {
      "auxiliary_loss_clip": 0.01156316,
      "auxiliary_loss_mlp": 0.00901293,
      "balance_loss_clip": 1.04826927,
      "balance_loss_mlp": 1.000633,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 1.7088448855137535,
      "language_loss": 0.65018058,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67075664,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.8554413318634033
    },
    {
      "auxiliary_loss_clip": 0.01159665,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.04714131,
      "balance_loss_mlp": 1.01623154,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.0165930181012337,
      "language_loss": 0.80052066,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.8223561,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.7445783615112305
    },
    {
      "auxiliary_loss_clip": 0.01068592,
      "auxiliary_loss_mlp": 0.01003056,
      "balance_loss_clip": 1.02076745,
      "balance_loss_mlp": 1.00187576,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.9021770682431647,
      "language_loss": 0.61961442,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64033091,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.1263906955718994
    },
    {
      "auxiliary_loss_clip": 0.01157337,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.04835415,
      "balance_loss_mlp": 1.01670301,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.8584752125732351,
      "language_loss": 0.79068458,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.8125025,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.7012081146240234
    },
    {
      "auxiliary_loss_clip": 0.01158867,
      "auxiliary_loss_mlp": 0.0102749,
      "balance_loss_clip": 1.05110717,
      "balance_loss_mlp": 1.02022421,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 15.096754727725846,
      "language_loss": 0.752653,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77451652,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.791689395904541
    },
    {
      "auxiliary_loss_clip": 0.01184284,
      "auxiliary_loss_mlp": 0.01032069,
      "balance_loss_clip": 1.0531373,
      "balance_loss_mlp": 1.02365279,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.7684530061697306,
      "language_loss": 0.73283565,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75499922,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.649111032485962
    },
    {
      "auxiliary_loss_clip": 0.01183984,
      "auxiliary_loss_mlp": 0.00901166,
      "balance_loss_clip": 1.05621171,
      "balance_loss_mlp": 1.00055814,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 1.8218416520230143,
      "language_loss": 0.74157792,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.76242942,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 3.5518980026245117
    },
    {
      "auxiliary_loss_clip": 0.01164356,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.05110383,
      "balance_loss_mlp": 1.02367878,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 2.37515056820564,
      "language_loss": 0.75985289,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.78180873,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.7061290740966797
    },
    {
      "auxiliary_loss_clip": 0.01162795,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.0495739,
      "balance_loss_mlp": 1.01886916,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.3403239373614966,
      "language_loss": 0.76421821,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78612077,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.692157745361328
    },
    {
      "auxiliary_loss_clip": 0.01183914,
      "auxiliary_loss_mlp": 0.01024889,
      "balance_loss_clip": 1.05437255,
      "balance_loss_mlp": 1.01668167,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 2.055149369230098,
      "language_loss": 0.78801632,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.81010437,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.724093198776245
    },
    {
      "auxiliary_loss_clip": 0.01183152,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.05494595,
      "balance_loss_mlp": 1.01810706,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.6963379217709844,
      "language_loss": 0.70931721,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.7314046,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.6712825298309326
    },
    {
      "auxiliary_loss_clip": 0.01162462,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.05243254,
      "balance_loss_mlp": 1.01707506,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.777861103143673,
      "language_loss": 0.7910558,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81293255,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.6977555751800537
    },
    {
      "auxiliary_loss_clip": 0.0114174,
      "auxiliary_loss_mlp": 0.01021276,
      "balance_loss_clip": 1.04686928,
      "balance_loss_mlp": 1.01315176,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.8392146949390653,
      "language_loss": 0.80615759,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.82778776,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 3.6698081493377686
    },
    {
      "auxiliary_loss_clip": 0.01161285,
      "auxiliary_loss_mlp": 0.01032352,
      "balance_loss_clip": 1.05227113,
      "balance_loss_mlp": 1.02421033,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.8851655833291974,
      "language_loss": 0.79163313,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81356955,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.7007312774658203
    },
    {
      "auxiliary_loss_clip": 0.01165316,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.05311537,
      "balance_loss_mlp": 1.02294493,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.915032241306869,
      "language_loss": 0.72988498,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75185561,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.6547954082489014
    },
    {
      "auxiliary_loss_clip": 0.01148571,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.05063212,
      "balance_loss_mlp": 1.01713526,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 1.9381916689192955,
      "language_loss": 0.71482688,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73656738,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 3.643986463546753
    },
    {
      "auxiliary_loss_clip": 0.01149539,
      "auxiliary_loss_mlp": 0.01026867,
      "balance_loss_clip": 1.05036092,
      "balance_loss_mlp": 1.01882052,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.9580154814177564,
      "language_loss": 0.61534572,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63710976,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.794151782989502
    },
    {
      "auxiliary_loss_clip": 0.01183943,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.05483294,
      "balance_loss_mlp": 1.01973069,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 2.1338809964243706,
      "language_loss": 0.72720343,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74932027,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.662445545196533
    },
    {
      "auxiliary_loss_clip": 0.01169479,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.0555743,
      "balance_loss_mlp": 1.02260709,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.7602249599132618,
      "language_loss": 0.78792572,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80992395,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.614508628845215
    },
    {
      "auxiliary_loss_clip": 0.01157187,
      "auxiliary_loss_mlp": 0.01032578,
      "balance_loss_clip": 1.05088925,
      "balance_loss_mlp": 1.02493095,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 2.195626648682648,
      "language_loss": 0.67841566,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.70031333,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.9511218070983887
    },
    {
      "auxiliary_loss_clip": 0.01170058,
      "auxiliary_loss_mlp": 0.0102698,
      "balance_loss_clip": 1.05239248,
      "balance_loss_mlp": 1.01907063,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.8467845919861656,
      "language_loss": 0.7150209,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73699129,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 3.614811897277832
    },
    {
      "auxiliary_loss_clip": 0.01148319,
      "auxiliary_loss_mlp": 0.01026835,
      "balance_loss_clip": 1.04722476,
      "balance_loss_mlp": 1.01831198,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 2.331908056870645,
      "language_loss": 0.76850778,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.7902593,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.735626459121704
    },
    {
      "auxiliary_loss_clip": 0.0118507,
      "auxiliary_loss_mlp": 0.01024692,
      "balance_loss_clip": 1.05630434,
      "balance_loss_mlp": 1.01619792,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.8709284416998846,
      "language_loss": 0.73735321,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75945079,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.6156833171844482
    },
    {
      "auxiliary_loss_clip": 0.01164049,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.04807639,
      "balance_loss_mlp": 1.01964998,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.3886317161363424,
      "language_loss": 0.74113703,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76305604,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.66339111328125
    },
    {
      "auxiliary_loss_clip": 0.01156685,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.04904723,
      "balance_loss_mlp": 1.02001154,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.4643571654317924,
      "language_loss": 0.83333325,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85518032,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.677495241165161
    },
    {
      "auxiliary_loss_clip": 0.01160908,
      "auxiliary_loss_mlp": 0.01027211,
      "balance_loss_clip": 1.05085897,
      "balance_loss_mlp": 1.01931977,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.7972417867283368,
      "language_loss": 0.74164855,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76352978,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.746117115020752
    },
    {
      "auxiliary_loss_clip": 0.01185565,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.05486763,
      "balance_loss_mlp": 1.01906085,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 2.228198748820464,
      "language_loss": 0.71798587,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.74010926,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.633000612258911
    },
    {
      "auxiliary_loss_clip": 0.0116268,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.05282581,
      "balance_loss_mlp": 1.01689243,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 2.621812496365529,
      "language_loss": 0.74700439,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.76888388,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.8106331825256348
    },
    {
      "auxiliary_loss_clip": 0.01181215,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 1.05310369,
      "balance_loss_mlp": 1.0223037,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 2.1840217185094515,
      "language_loss": 0.76071227,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78283203,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.6776504516601562
    },
    {
      "auxiliary_loss_clip": 0.01156749,
      "auxiliary_loss_mlp": 0.01025232,
      "balance_loss_clip": 1.04951203,
      "balance_loss_mlp": 1.01678586,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.844857259098991,
      "language_loss": 0.68694127,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70876104,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.613793134689331
    },
    {
      "auxiliary_loss_clip": 0.01188183,
      "auxiliary_loss_mlp": 0.01025178,
      "balance_loss_clip": 1.05852532,
      "balance_loss_mlp": 1.0174706,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 2.20285627866147,
      "language_loss": 0.81525117,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83738482,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.6314311027526855
    },
    {
      "auxiliary_loss_clip": 0.01168428,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.05088222,
      "balance_loss_mlp": 1.01595998,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 3.2651874647224455,
      "language_loss": 0.67488086,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69680893,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.70233154296875
    },
    {
      "auxiliary_loss_clip": 0.0116635,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 1.05488169,
      "balance_loss_mlp": 1.02156711,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.8591331937334645,
      "language_loss": 0.7998246,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82178408,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.6512532234191895
    },
    {
      "auxiliary_loss_clip": 0.01161496,
      "auxiliary_loss_mlp": 0.01029396,
      "balance_loss_clip": 1.05071962,
      "balance_loss_mlp": 1.02078903,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.0089032451846363,
      "language_loss": 0.7705723,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79248118,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 3.720479965209961
    },
    {
      "auxiliary_loss_clip": 0.01181589,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.05603838,
      "balance_loss_mlp": 1.01971436,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 1.8962129988408771,
      "language_loss": 0.76605451,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78815174,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.5961434841156006
    },
    {
      "auxiliary_loss_clip": 0.01157747,
      "auxiliary_loss_mlp": 0.01035277,
      "balance_loss_clip": 1.05309701,
      "balance_loss_mlp": 1.02751064,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.7798036875514325,
      "language_loss": 0.78953642,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81146669,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.6413214206695557
    },
    {
      "auxiliary_loss_clip": 0.01143805,
      "auxiliary_loss_mlp": 0.01023887,
      "balance_loss_clip": 1.04723239,
      "balance_loss_mlp": 1.01591206,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 1.8924015055376426,
      "language_loss": 0.77144861,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79312551,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.7018773555755615
    },
    {
      "auxiliary_loss_clip": 0.01177331,
      "auxiliary_loss_mlp": 0.01027663,
      "balance_loss_clip": 1.05613124,
      "balance_loss_mlp": 1.01938963,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 3.3538137483324904,
      "language_loss": 0.7602303,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78228033,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.662687301635742
    },
    {
      "auxiliary_loss_clip": 0.01169026,
      "auxiliary_loss_mlp": 0.0103288,
      "balance_loss_clip": 1.05263317,
      "balance_loss_mlp": 1.02528894,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 2.6332504740343476,
      "language_loss": 0.78739423,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.80941331,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.7468483448028564
    },
    {
      "auxiliary_loss_clip": 0.01175532,
      "auxiliary_loss_mlp": 0.01025514,
      "balance_loss_clip": 1.05403757,
      "balance_loss_mlp": 1.01696634,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.569056213390081,
      "language_loss": 0.76260638,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78461683,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 3.740835428237915
    },
    {
      "auxiliary_loss_clip": 0.01165395,
      "auxiliary_loss_mlp": 0.01024403,
      "balance_loss_clip": 1.05014873,
      "balance_loss_mlp": 1.01639819,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 2.5766399344106445,
      "language_loss": 0.84620422,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86810219,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.6794211864471436
    },
    {
      "auxiliary_loss_clip": 0.01174947,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.05694282,
      "balance_loss_mlp": 1.01551104,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 1.838581314266332,
      "language_loss": 0.82422364,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.84621525,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 3.463108539581299
    },
    {
      "auxiliary_loss_clip": 0.01173387,
      "auxiliary_loss_mlp": 0.00902182,
      "balance_loss_clip": 1.05432427,
      "balance_loss_mlp": 1.00074899,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.7597993301628256,
      "language_loss": 0.69601542,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71677113,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 2.628103017807007
    },
    {
      "auxiliary_loss_clip": 0.01160427,
      "auxiliary_loss_mlp": 0.01024531,
      "balance_loss_clip": 1.05460334,
      "balance_loss_mlp": 1.01643085,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 2.9477636051939764,
      "language_loss": 0.73941284,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76126242,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.7118144035339355
    },
    {
      "auxiliary_loss_clip": 0.01186388,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 1.05747247,
      "balance_loss_mlp": 1.01837206,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 5.447078520698097,
      "language_loss": 0.7255646,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.74768764,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.5283517837524414
    },
    {
      "auxiliary_loss_clip": 0.01178002,
      "auxiliary_loss_mlp": 0.0102484,
      "balance_loss_clip": 1.05463648,
      "balance_loss_mlp": 1.01681173,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.998939358332099,
      "language_loss": 0.79389077,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81591916,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.597456693649292
    },
    {
      "auxiliary_loss_clip": 0.01171238,
      "auxiliary_loss_mlp": 0.00901468,
      "balance_loss_clip": 1.05175161,
      "balance_loss_mlp": 1.00055039,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 2.6667271116190854,
      "language_loss": 0.81821173,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.83893877,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 3.5817043781280518
    },
    {
      "auxiliary_loss_clip": 0.01183216,
      "auxiliary_loss_mlp": 0.01023971,
      "balance_loss_clip": 1.05545771,
      "balance_loss_mlp": 1.01587677,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 2.8527801465427807,
      "language_loss": 0.82386625,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84593815,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.5902721881866455
    },
    {
      "auxiliary_loss_clip": 0.01184414,
      "auxiliary_loss_mlp": 0.01025066,
      "balance_loss_clip": 1.05712509,
      "balance_loss_mlp": 1.0170908,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 1.853285018369794,
      "language_loss": 0.63259697,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65469182,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.5764729976654053
    },
    {
      "auxiliary_loss_clip": 0.01173567,
      "auxiliary_loss_mlp": 0.01027054,
      "balance_loss_clip": 1.05358076,
      "balance_loss_mlp": 1.0195322,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.6685500511687774,
      "language_loss": 0.69682395,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.71883017,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.616473913192749
    },
    {
      "auxiliary_loss_clip": 0.01187903,
      "auxiliary_loss_mlp": 0.00901499,
      "balance_loss_clip": 1.05806756,
      "balance_loss_mlp": 1.0005759,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 2.2429369952246367,
      "language_loss": 0.675309,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.69620305,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.5621540546417236
    },
    {
      "auxiliary_loss_clip": 0.01149488,
      "auxiliary_loss_mlp": 0.01030515,
      "balance_loss_clip": 1.04944313,
      "balance_loss_mlp": 1.0224328,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 3.1639534156372084,
      "language_loss": 0.7827642,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80456424,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.8642568588256836
    },
    {
      "auxiliary_loss_clip": 0.01071787,
      "auxiliary_loss_mlp": 0.01004995,
      "balance_loss_clip": 1.0200758,
      "balance_loss_mlp": 1.00389791,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8024268655785062,
      "language_loss": 0.56569767,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58646548,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.2282416820526123
    },
    {
      "auxiliary_loss_clip": 0.01173979,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.05442941,
      "balance_loss_mlp": 1.01932263,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 2.429461473543503,
      "language_loss": 0.76649666,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78851402,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.6899423599243164
    },
    {
      "auxiliary_loss_clip": 0.01154403,
      "auxiliary_loss_mlp": 0.01026246,
      "balance_loss_clip": 1.05129313,
      "balance_loss_mlp": 1.0184083,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 2.760459214104385,
      "language_loss": 0.79090416,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.8127107,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.6860241889953613
    },
    {
      "auxiliary_loss_clip": 0.01161616,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.05000424,
      "balance_loss_mlp": 1.01896572,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 2.224570879630373,
      "language_loss": 0.79791081,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.81980091,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.671015501022339
    },
    {
      "auxiliary_loss_clip": 0.01176498,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.05694127,
      "balance_loss_mlp": 1.02065468,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 2.2053713845447236,
      "language_loss": 0.71287882,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73493391,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.6755216121673584
    },
    {
      "auxiliary_loss_clip": 0.01168682,
      "auxiliary_loss_mlp": 0.01031208,
      "balance_loss_clip": 1.05464149,
      "balance_loss_mlp": 1.02223182,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.5384152007577983,
      "language_loss": 0.72175145,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74375033,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.7203545570373535
    },
    {
      "auxiliary_loss_clip": 0.01168648,
      "auxiliary_loss_mlp": 0.0102806,
      "balance_loss_clip": 1.05053353,
      "balance_loss_mlp": 1.01941097,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 1.7566118888765718,
      "language_loss": 0.71611691,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73808396,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.6583497524261475
    },
    {
      "auxiliary_loss_clip": 0.01185837,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.05597317,
      "balance_loss_mlp": 1.0204432,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 1.9067380679421195,
      "language_loss": 0.84521544,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86736196,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 3.564093828201294
    },
    {
      "auxiliary_loss_clip": 0.01166141,
      "auxiliary_loss_mlp": 0.01027331,
      "balance_loss_clip": 1.04646325,
      "balance_loss_mlp": 1.01909924,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 1.9869934373079625,
      "language_loss": 0.67950678,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70144153,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.6174731254577637
    },
    {
      "auxiliary_loss_clip": 0.01078587,
      "auxiliary_loss_mlp": 0.01002511,
      "balance_loss_clip": 1.02188575,
      "balance_loss_mlp": 1.00141418,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7931977648641928,
      "language_loss": 0.5964964,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61730736,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.3250749111175537
    },
    {
      "auxiliary_loss_clip": 0.01171732,
      "auxiliary_loss_mlp": 0.01026403,
      "balance_loss_clip": 1.05269647,
      "balance_loss_mlp": 1.01792097,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.4713445889573618,
      "language_loss": 0.86720836,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88918972,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.628382444381714
    },
    {
      "auxiliary_loss_clip": 0.01159109,
      "auxiliary_loss_mlp": 0.0103318,
      "balance_loss_clip": 1.0520525,
      "balance_loss_mlp": 1.0251869,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 2.3262842397317427,
      "language_loss": 0.72900563,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75092852,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.706263542175293
    },
    {
      "auxiliary_loss_clip": 0.01180226,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.05330372,
      "balance_loss_mlp": 1.02147663,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 1.799214388592083,
      "language_loss": 0.61535549,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63745582,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.66178560256958
    },
    {
      "auxiliary_loss_clip": 0.0117669,
      "auxiliary_loss_mlp": 0.01025731,
      "balance_loss_clip": 1.05630445,
      "balance_loss_mlp": 1.01742792,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.2508260367084083,
      "language_loss": 0.75620073,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77822495,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.777005910873413
    },
    {
      "auxiliary_loss_clip": 0.01173002,
      "auxiliary_loss_mlp": 0.01033285,
      "balance_loss_clip": 1.05061448,
      "balance_loss_mlp": 1.02490437,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 1.9624176259690052,
      "language_loss": 0.71808171,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.74014455,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 3.6984140872955322
    },
    {
      "auxiliary_loss_clip": 0.0115943,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.05181086,
      "balance_loss_mlp": 1.0194459,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.3263268729514666,
      "language_loss": 0.73453903,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75641096,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.6278128623962402
    },
    {
      "auxiliary_loss_clip": 0.01188691,
      "auxiliary_loss_mlp": 0.01028402,
      "balance_loss_clip": 1.05797744,
      "balance_loss_mlp": 1.01981318,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 2.4350346867377706,
      "language_loss": 0.84602427,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86819518,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 3.5442769527435303
    },
    {
      "auxiliary_loss_clip": 0.0118059,
      "auxiliary_loss_mlp": 0.01030555,
      "balance_loss_clip": 1.05632508,
      "balance_loss_mlp": 1.0219419,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 1.9670761507258505,
      "language_loss": 0.82181334,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84392482,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.576810836791992
    },
    {
      "auxiliary_loss_clip": 0.01186251,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 1.05708671,
      "balance_loss_mlp": 1.02078414,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 2.3332032336260142,
      "language_loss": 0.62719071,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64934266,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.647989273071289
    },
    {
      "auxiliary_loss_clip": 0.01175337,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.05378759,
      "balance_loss_mlp": 1.01908112,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.7551582313227776,
      "language_loss": 0.80388856,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82590771,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.6099395751953125
    },
    {
      "auxiliary_loss_clip": 0.01169863,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.05567753,
      "balance_loss_mlp": 1.0182271,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 3.076794075782201,
      "language_loss": 0.69348013,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.71544778,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.637720823287964
    },
    {
      "auxiliary_loss_clip": 0.01068517,
      "auxiliary_loss_mlp": 0.01002714,
      "balance_loss_clip": 1.02165163,
      "balance_loss_mlp": 1.00165868,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.8982431552772389,
      "language_loss": 0.60985422,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63056648,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 4.2011027336120605
    },
    {
      "auxiliary_loss_clip": 0.01161029,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.04930639,
      "balance_loss_mlp": 1.0181601,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.9760898870645718,
      "language_loss": 0.74490738,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76677871,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.707845687866211
    },
    {
      "auxiliary_loss_clip": 0.01163459,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.05293059,
      "balance_loss_mlp": 1.01800907,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.225102055514919,
      "language_loss": 0.85510135,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87699872,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.678636312484741
    },
    {
      "auxiliary_loss_clip": 0.01166782,
      "auxiliary_loss_mlp": 0.01025415,
      "balance_loss_clip": 1.05507231,
      "balance_loss_mlp": 1.0176779,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.5662584668664798,
      "language_loss": 0.76631117,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.7882331,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.688854694366455
    },
    {
      "auxiliary_loss_clip": 0.01074926,
      "auxiliary_loss_mlp": 0.01003611,
      "balance_loss_clip": 1.0221076,
      "balance_loss_mlp": 1.00256777,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.725566991484042,
      "language_loss": 0.54730356,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56808895,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.082414388656616
    },
    {
      "auxiliary_loss_clip": 0.01172553,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.05219746,
      "balance_loss_mlp": 1.01878643,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.1847952197931604,
      "language_loss": 0.79188353,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81387627,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.6963698863983154
    },
    {
      "auxiliary_loss_clip": 0.01175736,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.05368567,
      "balance_loss_mlp": 1.01953435,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 2.140665160559506,
      "language_loss": 0.77734482,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79937792,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.635256290435791
    },
    {
      "auxiliary_loss_clip": 0.0117081,
      "auxiliary_loss_mlp": 0.01022664,
      "balance_loss_clip": 1.05108976,
      "balance_loss_mlp": 1.01505208,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.4972679222953325,
      "language_loss": 0.89134538,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91328013,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.646652936935425
    },
    {
      "auxiliary_loss_clip": 0.01072174,
      "auxiliary_loss_mlp": 0.01003084,
      "balance_loss_clip": 1.0218513,
      "balance_loss_mlp": 1.00191593,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9010089049401597,
      "language_loss": 0.63245308,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65320563,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.208397626876831
    },
    {
      "auxiliary_loss_clip": 0.01168244,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.04997051,
      "balance_loss_mlp": 1.02238595,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.5966848406500223,
      "language_loss": 0.78788149,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.80987316,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.6451971530914307
    },
    {
      "auxiliary_loss_clip": 0.01173045,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.05232203,
      "balance_loss_mlp": 1.02043223,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 3.449364866051064,
      "language_loss": 0.81128526,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83329892,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.625791072845459
    },
    {
      "auxiliary_loss_clip": 0.01173466,
      "auxiliary_loss_mlp": 0.01025305,
      "balance_loss_clip": 1.05068481,
      "balance_loss_mlp": 1.01723492,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 2.3138095764895263,
      "language_loss": 0.72708189,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.74906957,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.6797120571136475
    },
    {
      "auxiliary_loss_clip": 0.01152303,
      "auxiliary_loss_mlp": 0.01022757,
      "balance_loss_clip": 1.04564691,
      "balance_loss_mlp": 1.01538086,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 3.4718431264800764,
      "language_loss": 0.69682139,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71857202,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 3.7193005084991455
    },
    {
      "auxiliary_loss_clip": 0.0117903,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.05688739,
      "balance_loss_mlp": 1.01955032,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.6900873090078974,
      "language_loss": 0.77573609,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79779601,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.6319801807403564
    },
    {
      "auxiliary_loss_clip": 0.01151663,
      "auxiliary_loss_mlp": 0.010223,
      "balance_loss_clip": 1.04722691,
      "balance_loss_mlp": 1.0142442,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.4789268944850837,
      "language_loss": 0.82461894,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84635854,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.69340443611145
    },
    {
      "auxiliary_loss_clip": 0.01174413,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.05330455,
      "balance_loss_mlp": 1.0223639,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.1910450462517694,
      "language_loss": 0.77733767,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.79938865,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.6433939933776855
    },
    {
      "auxiliary_loss_clip": 0.0116248,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.05416667,
      "balance_loss_mlp": 1.01714635,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 3.22042231131201,
      "language_loss": 0.71288353,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73475075,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.67704176902771
    },
    {
      "auxiliary_loss_clip": 0.01186303,
      "auxiliary_loss_mlp": 0.01024904,
      "balance_loss_clip": 1.05552912,
      "balance_loss_mlp": 1.01631534,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.8907474146950227,
      "language_loss": 0.83712327,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.85923535,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.561861991882324
    },
    {
      "auxiliary_loss_clip": 0.01176926,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.0526216,
      "balance_loss_mlp": 1.0219022,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.9419343808251273,
      "language_loss": 0.82050657,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84257078,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.6705946922302246
    },
    {
      "auxiliary_loss_clip": 0.01184106,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.05473948,
      "balance_loss_mlp": 1.02330434,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 1.9306963178826912,
      "language_loss": 0.8095386,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83169496,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 3.437863826751709
    },
    {
      "auxiliary_loss_clip": 0.01147517,
      "auxiliary_loss_mlp": 0.01030584,
      "balance_loss_clip": 1.0476222,
      "balance_loss_mlp": 1.02183402,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 2.472360037526833,
      "language_loss": 0.66843343,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.6902144,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.7808475494384766
    },
    {
      "auxiliary_loss_clip": 0.01171185,
      "auxiliary_loss_mlp": 0.00900946,
      "balance_loss_clip": 1.05173135,
      "balance_loss_mlp": 1.00071466,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.726332926934332,
      "language_loss": 0.75870425,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.7794255,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.6573872566223145
    },
    {
      "auxiliary_loss_clip": 0.01171671,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.05499148,
      "balance_loss_mlp": 1.01935625,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 3.3451956096785054,
      "language_loss": 0.82116914,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84315223,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 3.6609156131744385
    },
    {
      "auxiliary_loss_clip": 0.01170569,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.05409288,
      "balance_loss_mlp": 1.02084374,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.6083636662108647,
      "language_loss": 0.74473429,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.7667253,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.738258123397827
    },
    {
      "auxiliary_loss_clip": 0.01167511,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.05261338,
      "balance_loss_mlp": 1.01977968,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 2.2734205638138767,
      "language_loss": 0.7844156,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80636549,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.6802237033843994
    },
    {
      "auxiliary_loss_clip": 0.01138058,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.04617119,
      "balance_loss_mlp": 1.01467896,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.9053874214833761,
      "language_loss": 0.75940001,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78100371,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.8345351219177246
    },
    {
      "auxiliary_loss_clip": 0.01172055,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.05392885,
      "balance_loss_mlp": 1.01813602,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 4.66586852471108,
      "language_loss": 0.72861075,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75059444,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.6496856212615967
    },
    {
      "auxiliary_loss_clip": 0.0115915,
      "auxiliary_loss_mlp": 0.01030988,
      "balance_loss_clip": 1.05317152,
      "balance_loss_mlp": 1.02291775,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 1.899358242247843,
      "language_loss": 0.92317307,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94507444,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 3.5580251216888428
    },
    {
      "auxiliary_loss_clip": 0.01155004,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.0500989,
      "balance_loss_mlp": 1.02039409,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.225100176652974,
      "language_loss": 0.66439569,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68622398,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.6803109645843506
    },
    {
      "auxiliary_loss_clip": 0.01145006,
      "auxiliary_loss_mlp": 0.01031919,
      "balance_loss_clip": 1.0440731,
      "balance_loss_mlp": 1.02351427,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 2.0530299753643977,
      "language_loss": 0.76277715,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78454638,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.7443997859954834
    },
    {
      "auxiliary_loss_clip": 0.01156113,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.0494597,
      "balance_loss_mlp": 1.02119231,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 1.8189681070381678,
      "language_loss": 0.7943809,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.8162322,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.6626503467559814
    },
    {
      "auxiliary_loss_clip": 0.01178935,
      "auxiliary_loss_mlp": 0.01026703,
      "balance_loss_clip": 1.05236053,
      "balance_loss_mlp": 1.01904416,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 10.402251186399365,
      "language_loss": 0.72819006,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75024647,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.5909993648529053
    },
    {
      "auxiliary_loss_clip": 0.01171112,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.05324447,
      "balance_loss_mlp": 1.01735473,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 1.9967354929501484,
      "language_loss": 0.72115541,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74312276,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.7045962810516357
    },
    {
      "auxiliary_loss_clip": 0.01159924,
      "auxiliary_loss_mlp": 0.01031099,
      "balance_loss_clip": 1.05045724,
      "balance_loss_mlp": 1.02346921,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.6169982949588995,
      "language_loss": 0.84916365,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87107384,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.698819637298584
    },
    {
      "auxiliary_loss_clip": 0.01183074,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.05284905,
      "balance_loss_mlp": 1.0206368,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.652562879732425,
      "language_loss": 0.75288355,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77500129,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.646456241607666
    },
    {
      "auxiliary_loss_clip": 0.01161976,
      "auxiliary_loss_mlp": 0.01027078,
      "balance_loss_clip": 1.05010724,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.9021521889069555,
      "language_loss": 0.64755058,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66944116,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.7455697059631348
    },
    {
      "auxiliary_loss_clip": 0.01145068,
      "auxiliary_loss_mlp": 0.01031018,
      "balance_loss_clip": 1.04619992,
      "balance_loss_mlp": 1.02313256,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 1.9322630994748182,
      "language_loss": 0.76326358,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.7850244,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.8100812435150146
    },
    {
      "auxiliary_loss_clip": 0.01168074,
      "auxiliary_loss_mlp": 0.01024348,
      "balance_loss_clip": 1.05274057,
      "balance_loss_mlp": 1.01678371,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.00840330122027,
      "language_loss": 0.79821634,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.8201406,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.7090201377868652
    },
    {
      "auxiliary_loss_clip": 0.011733,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.05162561,
      "balance_loss_mlp": 1.02230358,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 2.0266926812937975,
      "language_loss": 0.7342,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.75623536,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.6222541332244873
    },
    {
      "auxiliary_loss_clip": 0.01161019,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.05079865,
      "balance_loss_mlp": 1.01983452,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 2.0848367217608814,
      "language_loss": 0.76902598,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79091263,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 3.6334733963012695
    },
    {
      "auxiliary_loss_clip": 0.01065825,
      "auxiliary_loss_mlp": 0.01006623,
      "balance_loss_clip": 1.0203532,
      "balance_loss_mlp": 1.00549078,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 1.2617812493460105,
      "language_loss": 0.59745157,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61817598,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.1818580627441406
    },
    {
      "auxiliary_loss_clip": 0.01171158,
      "auxiliary_loss_mlp": 0.01023161,
      "balance_loss_clip": 1.05128646,
      "balance_loss_mlp": 1.01496851,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 2.558665538412481,
      "language_loss": 0.77395213,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79589528,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.613759994506836
    },
    {
      "auxiliary_loss_clip": 0.01083561,
      "auxiliary_loss_mlp": 0.01001253,
      "balance_loss_clip": 1.02174306,
      "balance_loss_mlp": 1.00021541,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.8817532635919363,
      "language_loss": 0.68817127,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70901942,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.12103009223938
    },
    {
      "auxiliary_loss_clip": 0.01141451,
      "auxiliary_loss_mlp": 0.0102329,
      "balance_loss_clip": 1.04720306,
      "balance_loss_mlp": 1.01616108,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.680261232204173,
      "language_loss": 0.82771188,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84935933,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.6319751739501953
    },
    {
      "auxiliary_loss_clip": 0.01168207,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.0509336,
      "balance_loss_mlp": 1.02299619,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 5.41778115948525,
      "language_loss": 0.80616772,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.82816082,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.6332712173461914
    },
    {
      "auxiliary_loss_clip": 0.01166996,
      "auxiliary_loss_mlp": 0.01023499,
      "balance_loss_clip": 1.0506134,
      "balance_loss_mlp": 1.01594067,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 2.948799241280379,
      "language_loss": 0.79865342,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.82055831,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.6267223358154297
    },
    {
      "auxiliary_loss_clip": 0.01168626,
      "auxiliary_loss_mlp": 0.01023317,
      "balance_loss_clip": 1.05192506,
      "balance_loss_mlp": 1.01559782,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.8702040297416307,
      "language_loss": 0.82069898,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84261835,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 3.6056838035583496
    },
    {
      "auxiliary_loss_clip": 0.01147235,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.05104971,
      "balance_loss_mlp": 1.01951385,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.7471043850199903,
      "language_loss": 0.85412979,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.8758775,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.729435920715332
    },
    {
      "auxiliary_loss_clip": 0.01087427,
      "auxiliary_loss_mlp": 0.01002017,
      "balance_loss_clip": 1.0219897,
      "balance_loss_mlp": 1.00100946,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7403287063457477,
      "language_loss": 0.55734646,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57824093,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.1271910667419434
    },
    {
      "auxiliary_loss_clip": 0.01151359,
      "auxiliary_loss_mlp": 0.01032192,
      "balance_loss_clip": 1.04861462,
      "balance_loss_mlp": 1.02490234,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.8376200759696757,
      "language_loss": 0.65801817,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.67985368,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 3.6216797828674316
    },
    {
      "auxiliary_loss_clip": 0.01184114,
      "auxiliary_loss_mlp": 0.01030401,
      "balance_loss_clip": 1.05428553,
      "balance_loss_mlp": 1.02213359,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 1.790318740456292,
      "language_loss": 0.74374235,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76588744,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.665665626525879
    },
    {
      "auxiliary_loss_clip": 0.01134861,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.04764366,
      "balance_loss_mlp": 1.01803744,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.1442977507747516,
      "language_loss": 0.81502306,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.83663654,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.7771058082580566
    },
    {
      "auxiliary_loss_clip": 0.01145034,
      "auxiliary_loss_mlp": 0.01024131,
      "balance_loss_clip": 1.04337215,
      "balance_loss_mlp": 1.01609039,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 3.4573806569793013,
      "language_loss": 0.77688336,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.79857498,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.7335474491119385
    },
    {
      "auxiliary_loss_clip": 0.01173723,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.05392146,
      "balance_loss_mlp": 1.02018332,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.8201043572030073,
      "language_loss": 0.6849575,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70696634,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.6554644107818604
    },
    {
      "auxiliary_loss_clip": 0.01156106,
      "auxiliary_loss_mlp": 0.0102372,
      "balance_loss_clip": 1.04662943,
      "balance_loss_mlp": 1.01575351,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.367834262441884,
      "language_loss": 0.82905424,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85085249,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 3.6923577785491943
    },
    {
      "auxiliary_loss_clip": 0.01146809,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.04892588,
      "balance_loss_mlp": 1.02127969,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 3.0338261766679957,
      "language_loss": 0.78718501,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80894405,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.7137832641601562
    },
    {
      "auxiliary_loss_clip": 0.01150937,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.04736018,
      "balance_loss_mlp": 1.01605153,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 2.153576118524939,
      "language_loss": 0.78986716,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.81160903,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.7899978160858154
    },
    {
      "auxiliary_loss_clip": 0.01182427,
      "auxiliary_loss_mlp": 0.01024845,
      "balance_loss_clip": 1.05371857,
      "balance_loss_mlp": 1.01667905,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.365117794230732,
      "language_loss": 0.78829384,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81036657,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.5954415798187256
    },
    {
      "auxiliary_loss_clip": 0.01171942,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.0513792,
      "balance_loss_mlp": 1.01687253,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 3.137579245382616,
      "language_loss": 0.66707361,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.68903697,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.794177770614624
    },
    {
      "auxiliary_loss_clip": 0.0111804,
      "auxiliary_loss_mlp": 0.0102512,
      "balance_loss_clip": 1.04371428,
      "balance_loss_mlp": 1.01702821,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.9931440156844986,
      "language_loss": 0.7407937,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.76222527,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.759963035583496
    },
    {
      "auxiliary_loss_clip": 0.01147398,
      "auxiliary_loss_mlp": 0.01027639,
      "balance_loss_clip": 1.04512656,
      "balance_loss_mlp": 1.01972914,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.6790613230726,
      "language_loss": 0.76728368,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.78903413,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.7406911849975586
    },
    {
      "auxiliary_loss_clip": 0.01151781,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.04848981,
      "balance_loss_mlp": 1.01897156,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.7932592227626916,
      "language_loss": 0.71096492,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.73274791,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.6684868335723877
    },
    {
      "auxiliary_loss_clip": 0.01155028,
      "auxiliary_loss_mlp": 0.01029536,
      "balance_loss_clip": 1.04698658,
      "balance_loss_mlp": 1.0220319,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 2.3782247916684423,
      "language_loss": 0.69730532,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.71915102,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.652404308319092
    },
    {
      "auxiliary_loss_clip": 0.01179827,
      "auxiliary_loss_mlp": 0.01024294,
      "balance_loss_clip": 1.05436945,
      "balance_loss_mlp": 1.0171181,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.952974420464608,
      "language_loss": 0.72085899,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74290019,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.6571502685546875
    },
    {
      "auxiliary_loss_clip": 0.01169294,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.05004132,
      "balance_loss_mlp": 1.02034402,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.002080899501722,
      "language_loss": 0.82957834,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.85154998,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.598598003387451
    },
    {
      "auxiliary_loss_clip": 0.01164665,
      "auxiliary_loss_mlp": 0.01023701,
      "balance_loss_clip": 1.05151224,
      "balance_loss_mlp": 1.01647353,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 2.165543314483038,
      "language_loss": 0.83399737,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85588098,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.656287908554077
    },
    {
      "auxiliary_loss_clip": 0.01171497,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.05372071,
      "balance_loss_mlp": 1.01895249,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.8430237683597612,
      "language_loss": 0.71937376,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74134904,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 3.658137798309326
    },
    {
      "auxiliary_loss_clip": 0.01168675,
      "auxiliary_loss_mlp": 0.01022526,
      "balance_loss_clip": 1.05492437,
      "balance_loss_mlp": 1.01536453,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 2.450119026315678,
      "language_loss": 0.7264778,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74838984,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.7011470794677734
    },
    {
      "auxiliary_loss_clip": 0.0115526,
      "auxiliary_loss_mlp": 0.01024875,
      "balance_loss_clip": 1.05003786,
      "balance_loss_mlp": 1.01754403,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.5298346379743784,
      "language_loss": 0.74530137,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76710272,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.729194164276123
    },
    {
      "auxiliary_loss_clip": 0.01151897,
      "auxiliary_loss_mlp": 0.01030741,
      "balance_loss_clip": 1.05467081,
      "balance_loss_mlp": 1.02301371,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 2.4965351947996854,
      "language_loss": 0.73813862,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.759965,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.711662530899048
    },
    {
      "auxiliary_loss_clip": 0.0116115,
      "auxiliary_loss_mlp": 0.00900556,
      "balance_loss_clip": 1.05102372,
      "balance_loss_mlp": 1.00082779,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 3.673011648735606,
      "language_loss": 0.83890605,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.85952306,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.607816457748413
    },
    {
      "auxiliary_loss_clip": 0.01184036,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.05637383,
      "balance_loss_mlp": 1.02009058,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.746951979103663,
      "language_loss": 0.70515198,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72727102,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.618220090866089
    },
    {
      "auxiliary_loss_clip": 0.01071336,
      "auxiliary_loss_mlp": 0.01003206,
      "balance_loss_clip": 1.02548504,
      "balance_loss_mlp": 1.0022583,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.8173287848733412,
      "language_loss": 0.61927724,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64002264,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.278698205947876
    },
    {
      "auxiliary_loss_clip": 0.01169053,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 1.05068552,
      "balance_loss_mlp": 1.01687622,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.6742572534416573,
      "language_loss": 0.70899606,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.73093462,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 3.6108717918395996
    },
    {
      "auxiliary_loss_clip": 0.01165611,
      "auxiliary_loss_mlp": 0.01036331,
      "balance_loss_clip": 1.05153072,
      "balance_loss_mlp": 1.02872539,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.608590051809097,
      "language_loss": 0.78962755,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81164706,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.706979513168335
    },
    {
      "auxiliary_loss_clip": 0.01161907,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.05252063,
      "balance_loss_mlp": 1.01885331,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 1.9139133755486872,
      "language_loss": 0.73889554,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76077795,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.6880271434783936
    },
    {
      "auxiliary_loss_clip": 0.01140928,
      "auxiliary_loss_mlp": 0.01024996,
      "balance_loss_clip": 1.04918003,
      "balance_loss_mlp": 1.01762581,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.7621474156594457,
      "language_loss": 0.79680955,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.81846881,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 3.6931395530700684
    },
    {
      "auxiliary_loss_clip": 0.01162599,
      "auxiliary_loss_mlp": 0.01025187,
      "balance_loss_clip": 1.05143321,
      "balance_loss_mlp": 1.0172714,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.788022077241275,
      "language_loss": 0.70929337,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.73117125,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.6851444244384766
    },
    {
      "auxiliary_loss_clip": 0.01074711,
      "auxiliary_loss_mlp": 0.01003406,
      "balance_loss_clip": 1.02253938,
      "balance_loss_mlp": 1.00250006,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.808060246083932,
      "language_loss": 0.60327667,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62405789,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.2233574390411377
    },
    {
      "auxiliary_loss_clip": 0.0116392,
      "auxiliary_loss_mlp": 0.01022557,
      "balance_loss_clip": 1.05103028,
      "balance_loss_mlp": 1.0153389,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.6547543481415268,
      "language_loss": 0.75326657,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.7751314,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.6847152709960938
    },
    {
      "auxiliary_loss_clip": 0.01169337,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.05599475,
      "balance_loss_mlp": 1.01905727,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.946467412908849,
      "language_loss": 0.76530153,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78727007,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.7105512619018555
    },
    {
      "auxiliary_loss_clip": 0.01172976,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.05521703,
      "balance_loss_mlp": 1.01837516,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.112563314830739,
      "language_loss": 0.86177498,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88376045,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 3.5909948348999023
    },
    {
      "auxiliary_loss_clip": 0.01143439,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.04613709,
      "balance_loss_mlp": 1.01690912,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 2.073922237503995,
      "language_loss": 0.78508759,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80676687,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.7438831329345703
    },
    {
      "auxiliary_loss_clip": 0.01153844,
      "auxiliary_loss_mlp": 0.01024118,
      "balance_loss_clip": 1.05191469,
      "balance_loss_mlp": 1.01683462,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 3.0102145090621306,
      "language_loss": 0.75968307,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.78146267,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.632174491882324
    },
    {
      "auxiliary_loss_clip": 0.01146568,
      "auxiliary_loss_mlp": 0.01025863,
      "balance_loss_clip": 1.04691625,
      "balance_loss_mlp": 1.01806045,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 2.282574049933727,
      "language_loss": 0.68226695,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.70399129,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.831571578979492
    },
    {
      "auxiliary_loss_clip": 0.01162557,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.0565542,
      "balance_loss_mlp": 1.01781142,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 3.0974366940811073,
      "language_loss": 0.66488051,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68676752,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.756279230117798
    },
    {
      "auxiliary_loss_clip": 0.01169008,
      "auxiliary_loss_mlp": 0.01024044,
      "balance_loss_clip": 1.05110168,
      "balance_loss_mlp": 1.01662278,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 1.966819231329082,
      "language_loss": 0.67825359,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70018411,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.6357831954956055
    },
    {
      "auxiliary_loss_clip": 0.01172845,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.05277824,
      "balance_loss_mlp": 1.0164876,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 1.8153924789581877,
      "language_loss": 0.65413237,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67609781,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.7419188022613525
    },
    {
      "auxiliary_loss_clip": 0.01169158,
      "auxiliary_loss_mlp": 0.01027641,
      "balance_loss_clip": 1.05301571,
      "balance_loss_mlp": 1.01959467,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 3.751500635213468,
      "language_loss": 0.79429483,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81626284,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.6640403270721436
    },
    {
      "auxiliary_loss_clip": 0.01179012,
      "auxiliary_loss_mlp": 0.01032338,
      "balance_loss_clip": 1.05431843,
      "balance_loss_mlp": 1.02502453,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 1.9639327834636984,
      "language_loss": 0.84609783,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86821133,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.6187376976013184
    },
    {
      "auxiliary_loss_clip": 0.01144405,
      "auxiliary_loss_mlp": 0.01029093,
      "balance_loss_clip": 1.04982424,
      "balance_loss_mlp": 1.02189827,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 2.7487828515537793,
      "language_loss": 0.71109474,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73282969,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.7473649978637695
    },
    {
      "auxiliary_loss_clip": 0.01159719,
      "auxiliary_loss_mlp": 0.01027556,
      "balance_loss_clip": 1.05556464,
      "balance_loss_mlp": 1.01985538,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.7977566289944094,
      "language_loss": 0.57354081,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59541351,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.8567054271698
    },
    {
      "auxiliary_loss_clip": 0.01154006,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04741228,
      "balance_loss_mlp": 1.0180608,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 2.881247007546889,
      "language_loss": 0.87476915,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89656466,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 3.5706684589385986
    },
    {
      "auxiliary_loss_clip": 0.01161611,
      "auxiliary_loss_mlp": 0.0102719,
      "balance_loss_clip": 1.04937124,
      "balance_loss_mlp": 1.01887488,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.2694662623169894,
      "language_loss": 0.82177424,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.84366226,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.718360424041748
    },
    {
      "auxiliary_loss_clip": 0.01167094,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.05224252,
      "balance_loss_mlp": 1.02093983,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 2.083860899390913,
      "language_loss": 0.79322922,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.815184,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 2.6700551509857178
    },
    {
      "auxiliary_loss_clip": 0.01168826,
      "auxiliary_loss_mlp": 0.01023353,
      "balance_loss_clip": 1.05130744,
      "balance_loss_mlp": 1.01582789,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.7966164816614254,
      "language_loss": 0.85399723,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87591898,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.749185800552368
    },
    {
      "auxiliary_loss_clip": 0.01169285,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.05105329,
      "balance_loss_mlp": 1.01883566,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.8194730390088572,
      "language_loss": 0.70124501,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72319996,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.706681251525879
    },
    {
      "auxiliary_loss_clip": 0.01154476,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.05027223,
      "balance_loss_mlp": 1.02275443,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.7688812689193891,
      "language_loss": 0.73679912,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75864774,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.7862651348114014
    },
    {
      "auxiliary_loss_clip": 0.01163131,
      "auxiliary_loss_mlp": 0.01025162,
      "balance_loss_clip": 1.05215383,
      "balance_loss_mlp": 1.01741946,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.1302658112121144,
      "language_loss": 0.69417095,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71605396,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.786726713180542
    },
    {
      "auxiliary_loss_clip": 0.01174482,
      "auxiliary_loss_mlp": 0.01027617,
      "balance_loss_clip": 1.05601549,
      "balance_loss_mlp": 1.01974344,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 1.866892559580549,
      "language_loss": 0.72298104,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74500203,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 3.531219959259033
    },
    {
      "auxiliary_loss_clip": 0.01159288,
      "auxiliary_loss_mlp": 0.01026752,
      "balance_loss_clip": 1.05067563,
      "balance_loss_mlp": 1.01897955,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.002029323966826,
      "language_loss": 0.61661196,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.63847238,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.7764389514923096
    },
    {
      "auxiliary_loss_clip": 0.01157599,
      "auxiliary_loss_mlp": 0.01023361,
      "balance_loss_clip": 1.05541992,
      "balance_loss_mlp": 1.01571393,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 2.591227375690676,
      "language_loss": 0.77327776,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.7950874,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.747281074523926
    },
    {
      "auxiliary_loss_clip": 0.01171287,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.05414116,
      "balance_loss_mlp": 1.01882756,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.379940565231675,
      "language_loss": 0.73813075,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.76011145,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.6403887271881104
    },
    {
      "auxiliary_loss_clip": 0.0115771,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.05072963,
      "balance_loss_mlp": 1.02309132,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.6066087724567824,
      "language_loss": 0.78439289,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80627531,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 3.739743709564209
    },
    {
      "auxiliary_loss_clip": 0.01157154,
      "auxiliary_loss_mlp": 0.00900599,
      "balance_loss_clip": 1.04909623,
      "balance_loss_mlp": 1.00086856,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.174990954125356,
      "language_loss": 0.77742791,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.7980054,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.7212748527526855
    },
    {
      "auxiliary_loss_clip": 0.01155842,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.04966259,
      "balance_loss_mlp": 1.02093053,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.4194086737196785,
      "language_loss": 0.72109807,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.74294883,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.7388689517974854
    },
    {
      "auxiliary_loss_clip": 0.01150489,
      "auxiliary_loss_mlp": 0.01024524,
      "balance_loss_clip": 1.04721642,
      "balance_loss_mlp": 1.0174098,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 2.347100359101571,
      "language_loss": 0.86064804,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88239819,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.716763734817505
    },
    {
      "auxiliary_loss_clip": 0.01171905,
      "auxiliary_loss_mlp": 0.01030836,
      "balance_loss_clip": 1.0526365,
      "balance_loss_mlp": 1.02310562,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 2.685293816351724,
      "language_loss": 0.90980566,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.93183303,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.6202831268310547
    },
    {
      "auxiliary_loss_clip": 0.01175388,
      "auxiliary_loss_mlp": 0.01026446,
      "balance_loss_clip": 1.05179238,
      "balance_loss_mlp": 1.0183692,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.716533933418519,
      "language_loss": 0.81863439,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.8406527,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 3.6297028064727783
    },
    {
      "auxiliary_loss_clip": 0.01181582,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.05508351,
      "balance_loss_mlp": 1.01812863,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.8895110252758895,
      "language_loss": 0.87341738,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89549208,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.600210666656494
    },
    {
      "auxiliary_loss_clip": 0.01168143,
      "auxiliary_loss_mlp": 0.01023853,
      "balance_loss_clip": 1.05091619,
      "balance_loss_mlp": 1.01574063,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.8551944831493532,
      "language_loss": 0.85802883,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87994879,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.710407257080078
    },
    {
      "auxiliary_loss_clip": 0.01182402,
      "auxiliary_loss_mlp": 0.00900641,
      "balance_loss_clip": 1.05351675,
      "balance_loss_mlp": 1.00094664,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.7475823290796186,
      "language_loss": 0.81279415,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.8336246,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.5996291637420654
    },
    {
      "auxiliary_loss_clip": 0.01163943,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.04887795,
      "balance_loss_mlp": 1.01729727,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.7308998106796842,
      "language_loss": 0.75425804,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77614552,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.7564332485198975
    },
    {
      "auxiliary_loss_clip": 0.01175085,
      "auxiliary_loss_mlp": 0.01026156,
      "balance_loss_clip": 1.0532968,
      "balance_loss_mlp": 1.01807332,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.676195215265199,
      "language_loss": 0.83611161,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85812402,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.552008628845215
    },
    {
      "auxiliary_loss_clip": 0.0117595,
      "auxiliary_loss_mlp": 0.01025328,
      "balance_loss_clip": 1.05729365,
      "balance_loss_mlp": 1.01809227,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.772121836339095,
      "language_loss": 0.73758686,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75959963,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.6821534633636475
    },
    {
      "auxiliary_loss_clip": 0.01142301,
      "auxiliary_loss_mlp": 0.01029402,
      "balance_loss_clip": 1.04633522,
      "balance_loss_mlp": 1.02112257,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.4450313419429355,
      "language_loss": 0.79103816,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.81275511,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.773237943649292
    },
    {
      "auxiliary_loss_clip": 0.01165482,
      "auxiliary_loss_mlp": 0.01028921,
      "balance_loss_clip": 1.05545783,
      "balance_loss_mlp": 1.02079737,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 1.6946182574629844,
      "language_loss": 0.63346064,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65540469,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.7562012672424316
    },
    {
      "auxiliary_loss_clip": 0.01143295,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.04372036,
      "balance_loss_mlp": 1.0174619,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 3.22648332430461,
      "language_loss": 0.86732829,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88901007,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.6723928451538086
    },
    {
      "auxiliary_loss_clip": 0.01171441,
      "auxiliary_loss_mlp": 0.01026161,
      "balance_loss_clip": 1.05414891,
      "balance_loss_mlp": 1.01897883,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.573111349053484,
      "language_loss": 0.74646497,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76844102,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 3.6511082649230957
    },
    {
      "auxiliary_loss_clip": 0.01174809,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.0518024,
      "balance_loss_mlp": 1.01859331,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 2.278720952731931,
      "language_loss": 0.77750516,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79952127,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.7320947647094727
    },
    {
      "auxiliary_loss_clip": 0.01163274,
      "auxiliary_loss_mlp": 0.01025535,
      "balance_loss_clip": 1.05140948,
      "balance_loss_mlp": 1.0182395,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.9986101350208134,
      "language_loss": 0.73802006,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.7599082,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.7451229095458984
    },
    {
      "auxiliary_loss_clip": 0.01172181,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.0541501,
      "balance_loss_mlp": 1.01747966,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.6159573353671182,
      "language_loss": 0.89358032,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91556096,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.7277510166168213
    },
    {
      "auxiliary_loss_clip": 0.0117257,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 1.05400836,
      "balance_loss_mlp": 1.0197767,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.607232967711564,
      "language_loss": 0.75372171,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77572167,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.649665355682373
    },
    {
      "auxiliary_loss_clip": 0.01071268,
      "auxiliary_loss_mlp": 0.00891245,
      "balance_loss_clip": 1.02127647,
      "balance_loss_mlp": 0.99998683,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7570496015347652,
      "language_loss": 0.57999539,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59962058,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.2989795207977295
    },
    {
      "auxiliary_loss_clip": 0.01171991,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.05243301,
      "balance_loss_mlp": 1.01631558,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 2.069651058768958,
      "language_loss": 0.68723029,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.70918679,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.768611192703247
    },
    {
      "auxiliary_loss_clip": 0.01165131,
      "auxiliary_loss_mlp": 0.01021413,
      "balance_loss_clip": 1.05183053,
      "balance_loss_mlp": 1.01412368,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.9986333561606175,
      "language_loss": 0.84519076,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86705613,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.5676751136779785
    },
    {
      "auxiliary_loss_clip": 0.0115412,
      "auxiliary_loss_mlp": 0.01020508,
      "balance_loss_clip": 1.05332792,
      "balance_loss_mlp": 1.01303637,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 2.4087635006464816,
      "language_loss": 0.73991567,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.76166195,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 3.6663594245910645
    },
    {
      "auxiliary_loss_clip": 0.01178266,
      "auxiliary_loss_mlp": 0.01021448,
      "balance_loss_clip": 1.05414796,
      "balance_loss_mlp": 1.01434875,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.102087331652461,
      "language_loss": 0.71443868,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73643577,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 2.565741777420044
    },
    {
      "auxiliary_loss_clip": 0.01176631,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.05152798,
      "balance_loss_mlp": 1.01795638,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.7080155024623012,
      "language_loss": 0.62036365,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.6423943,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.8442928791046143
    },
    {
      "auxiliary_loss_clip": 0.01149309,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04802918,
      "balance_loss_mlp": 1.01907909,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 2.197430364778502,
      "language_loss": 0.678002,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.69976181,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 3.684462785720825
    },
    {
      "auxiliary_loss_clip": 0.01183312,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.05460405,
      "balance_loss_mlp": 1.01948678,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 1.9246737336285293,
      "language_loss": 0.71686184,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73896444,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.6456735134124756
    },
    {
      "auxiliary_loss_clip": 0.01168294,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.05351675,
      "balance_loss_mlp": 1.02061522,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.4199523393273243,
      "language_loss": 0.80834728,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83031183,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.631688117980957
    },
    {
      "auxiliary_loss_clip": 0.01159272,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 1.0484271,
      "balance_loss_mlp": 1.01562107,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 2.42977078061792,
      "language_loss": 0.71131706,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73313898,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.633329391479492
    },
    {
      "auxiliary_loss_clip": 0.01171629,
      "auxiliary_loss_mlp": 0.01030703,
      "balance_loss_clip": 1.05160737,
      "balance_loss_mlp": 1.02347541,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.8641734728606634,
      "language_loss": 0.79892105,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.82094437,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.7130320072174072
    },
    {
      "auxiliary_loss_clip": 0.01159869,
      "auxiliary_loss_mlp": 0.01027342,
      "balance_loss_clip": 1.04994607,
      "balance_loss_mlp": 1.01976061,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 3.044539355670025,
      "language_loss": 0.90882355,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93069565,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.689208507537842
    },
    {
      "auxiliary_loss_clip": 0.01168141,
      "auxiliary_loss_mlp": 0.01026103,
      "balance_loss_clip": 1.0528903,
      "balance_loss_mlp": 1.01866996,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.479223195093148,
      "language_loss": 0.79641646,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.8183589,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 3.538616895675659
    },
    {
      "auxiliary_loss_clip": 0.01167679,
      "auxiliary_loss_mlp": 0.01027202,
      "balance_loss_clip": 1.05128264,
      "balance_loss_mlp": 1.01989448,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 3.7936604166546677,
      "language_loss": 0.83193278,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.8538816,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.6806833744049072
    },
    {
      "auxiliary_loss_clip": 0.01184915,
      "auxiliary_loss_mlp": 0.01029384,
      "balance_loss_clip": 1.05715156,
      "balance_loss_mlp": 1.02195692,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 4.1582854434548295,
      "language_loss": 0.83043873,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85258174,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.548313617706299
    },
    {
      "auxiliary_loss_clip": 0.01180891,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.05540907,
      "balance_loss_mlp": 1.02027535,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.1450812855993533,
      "language_loss": 0.79082358,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81291109,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.6975409984588623
    },
    {
      "auxiliary_loss_clip": 0.01170687,
      "auxiliary_loss_mlp": 0.01021401,
      "balance_loss_clip": 1.05349684,
      "balance_loss_mlp": 1.01439703,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.9094308568043865,
      "language_loss": 0.72365916,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74558002,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.585726261138916
    },
    {
      "auxiliary_loss_clip": 0.01157398,
      "auxiliary_loss_mlp": 0.00901776,
      "balance_loss_clip": 1.05234909,
      "balance_loss_mlp": 1.00088251,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.7757274228697542,
      "language_loss": 0.74403262,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76462442,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.935007095336914
    },
    {
      "auxiliary_loss_clip": 0.01171391,
      "auxiliary_loss_mlp": 0.01027306,
      "balance_loss_clip": 1.05382395,
      "balance_loss_mlp": 1.020046,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 2.045365107708614,
      "language_loss": 0.73647702,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.75846398,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.6084377765655518
    },
    {
      "auxiliary_loss_clip": 0.01160116,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.05280733,
      "balance_loss_mlp": 1.01608872,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 3.9250888669182187,
      "language_loss": 0.78266084,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80450022,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.6978530883789062
    },
    {
      "auxiliary_loss_clip": 0.01151928,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.04785585,
      "balance_loss_mlp": 1.02065206,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.1165494053215177,
      "language_loss": 0.71058351,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73239112,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.6400458812713623
    },
    {
      "auxiliary_loss_clip": 0.01165416,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.05325937,
      "balance_loss_mlp": 1.01886225,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 2.100173447021633,
      "language_loss": 0.77043867,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.79235882,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.706416606903076
    },
    {
      "auxiliary_loss_clip": 0.01184114,
      "auxiliary_loss_mlp": 0.01028594,
      "balance_loss_clip": 1.05770445,
      "balance_loss_mlp": 1.02100337,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 1.9622859731031859,
      "language_loss": 0.70340097,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72552806,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 3.5691721439361572
    },
    {
      "auxiliary_loss_clip": 0.01165415,
      "auxiliary_loss_mlp": 0.01020378,
      "balance_loss_clip": 1.04978049,
      "balance_loss_mlp": 1.01284361,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.8591031633337347,
      "language_loss": 0.78243101,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80428898,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.6953136920928955
    },
    {
      "auxiliary_loss_clip": 0.01150286,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.05224097,
      "balance_loss_mlp": 1.02121615,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.1364885713326665,
      "language_loss": 0.76447183,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78626609,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 2.7299373149871826
    },
    {
      "auxiliary_loss_clip": 0.0118269,
      "auxiliary_loss_mlp": 0.01026215,
      "balance_loss_clip": 1.05500758,
      "balance_loss_mlp": 1.01880074,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.1255150556053053,
      "language_loss": 0.75732982,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.77941895,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.6221466064453125
    },
    {
      "auxiliary_loss_clip": 0.01158127,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.04955482,
      "balance_loss_mlp": 1.02314544,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.7461460407896443,
      "language_loss": 0.70030981,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72219926,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.853410243988037
    },
    {
      "auxiliary_loss_clip": 0.01168804,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.05782175,
      "balance_loss_mlp": 1.01823795,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 1.6657174283422056,
      "language_loss": 0.76498455,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78692633,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.6744353771209717
    },
    {
      "auxiliary_loss_clip": 0.01174608,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.05546999,
      "balance_loss_mlp": 1.02033854,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.8012515772962447,
      "language_loss": 0.74941266,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.77143985,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.649799108505249
    },
    {
      "auxiliary_loss_clip": 0.0116583,
      "auxiliary_loss_mlp": 0.01023304,
      "balance_loss_clip": 1.05351758,
      "balance_loss_mlp": 1.01612759,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.9857293611048952,
      "language_loss": 0.73196036,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75385171,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.6524274349212646
    },
    {
      "auxiliary_loss_clip": 0.01169503,
      "auxiliary_loss_mlp": 0.01025179,
      "balance_loss_clip": 1.05409122,
      "balance_loss_mlp": 1.01817584,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.8898604714865497,
      "language_loss": 0.79929864,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82124543,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 3.643899917602539
    },
    {
      "auxiliary_loss_clip": 0.01179056,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05461645,
      "balance_loss_mlp": 1.01887965,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.5970962620978233,
      "language_loss": 0.7951777,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81723017,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.6367948055267334
    },
    {
      "auxiliary_loss_clip": 0.01157793,
      "auxiliary_loss_mlp": 0.01028937,
      "balance_loss_clip": 1.05026507,
      "balance_loss_mlp": 1.02138186,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.8404233903413105,
      "language_loss": 0.77226603,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79413337,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.64766001701355
    },
    {
      "auxiliary_loss_clip": 0.01151915,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.05224729,
      "balance_loss_mlp": 1.02039969,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.6346630169437337,
      "language_loss": 0.8092913,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83109003,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 3.625561475753784
    },
    {
      "auxiliary_loss_clip": 0.01171524,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.05393171,
      "balance_loss_mlp": 1.01751864,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.3327851583702355,
      "language_loss": 0.67901361,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70098424,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.593135356903076
    },
    {
      "auxiliary_loss_clip": 0.01167197,
      "auxiliary_loss_mlp": 0.01019759,
      "balance_loss_clip": 1.05269635,
      "balance_loss_mlp": 1.01234388,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.4160157697448847,
      "language_loss": 0.66316634,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68503594,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.678840160369873
    },
    {
      "auxiliary_loss_clip": 0.01157424,
      "auxiliary_loss_mlp": 0.00901166,
      "balance_loss_clip": 1.04793453,
      "balance_loss_mlp": 1.00085759,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.6945594091619536,
      "language_loss": 0.70671278,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72729862,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.7914838790893555
    },
    {
      "auxiliary_loss_clip": 0.01168644,
      "auxiliary_loss_mlp": 0.0102598,
      "balance_loss_clip": 1.05122256,
      "balance_loss_mlp": 1.01889288,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.8392376049361305,
      "language_loss": 0.81412435,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83607054,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.638770580291748
    },
    {
      "auxiliary_loss_clip": 0.01157141,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.05061781,
      "balance_loss_mlp": 1.01850724,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.8101788273237833,
      "language_loss": 0.78399628,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80582476,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.6838791370391846
    },
    {
      "auxiliary_loss_clip": 0.01157887,
      "auxiliary_loss_mlp": 0.01024925,
      "balance_loss_clip": 1.05035794,
      "balance_loss_mlp": 1.01768279,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.7213770218519144,
      "language_loss": 0.78063363,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80246174,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 3.749328851699829
    },
    {
      "auxiliary_loss_clip": 0.01166628,
      "auxiliary_loss_mlp": 0.01023324,
      "balance_loss_clip": 1.05147874,
      "balance_loss_mlp": 1.01530433,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 3.2602956515322408,
      "language_loss": 0.77662295,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.79852247,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.6955060958862305
    },
    {
      "auxiliary_loss_clip": 0.01171817,
      "auxiliary_loss_mlp": 0.01023439,
      "balance_loss_clip": 1.05329013,
      "balance_loss_mlp": 1.01606584,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 3.155874186303298,
      "language_loss": 0.72649968,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74845219,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.7181434631347656
    },
    {
      "auxiliary_loss_clip": 0.01177425,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05284715,
      "balance_loss_mlp": 1.01818681,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.438772464768719,
      "language_loss": 0.885934,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90796906,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.6023824214935303
    },
    {
      "auxiliary_loss_clip": 0.01165582,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.05322063,
      "balance_loss_mlp": 1.01654077,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.1184257843894194,
      "language_loss": 0.68362129,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70551562,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.7769978046417236
    },
    {
      "auxiliary_loss_clip": 0.01178881,
      "auxiliary_loss_mlp": 0.01026505,
      "balance_loss_clip": 1.05373394,
      "balance_loss_mlp": 1.01933503,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 2.2325056503073,
      "language_loss": 0.78690863,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.80896246,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.6394081115722656
    },
    {
      "auxiliary_loss_clip": 0.0115937,
      "auxiliary_loss_mlp": 0.01024884,
      "balance_loss_clip": 1.04952502,
      "balance_loss_mlp": 1.01757026,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 2.2500302606268323,
      "language_loss": 0.7416023,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.76344484,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.6894588470458984
    },
    {
      "auxiliary_loss_clip": 0.01160142,
      "auxiliary_loss_mlp": 0.01022953,
      "balance_loss_clip": 1.04963827,
      "balance_loss_mlp": 1.01575327,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.7306978326024849,
      "language_loss": 0.75348008,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77531105,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.696992874145508
    },
    {
      "auxiliary_loss_clip": 0.01159049,
      "auxiliary_loss_mlp": 0.01024954,
      "balance_loss_clip": 1.05004382,
      "balance_loss_mlp": 1.01724076,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.285020850279039,
      "language_loss": 0.73506612,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75690615,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.653237819671631
    },
    {
      "auxiliary_loss_clip": 0.01175282,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.05556178,
      "balance_loss_mlp": 1.02131295,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.678688154295661,
      "language_loss": 0.81421304,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83626342,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.6159470081329346
    },
    {
      "auxiliary_loss_clip": 0.01172061,
      "auxiliary_loss_mlp": 0.01022351,
      "balance_loss_clip": 1.05394924,
      "balance_loss_mlp": 1.01406014,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 2.1486318065549694,
      "language_loss": 0.61751616,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.63946021,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 3.7046635150909424
    },
    {
      "auxiliary_loss_clip": 0.01171872,
      "auxiliary_loss_mlp": 0.01023753,
      "balance_loss_clip": 1.05505705,
      "balance_loss_mlp": 1.01621294,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 2.1030801835340966,
      "language_loss": 0.88345158,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90540791,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 2.6897449493408203
    },
    {
      "auxiliary_loss_clip": 0.01171173,
      "auxiliary_loss_mlp": 0.01025287,
      "balance_loss_clip": 1.05475855,
      "balance_loss_mlp": 1.01807213,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 1.9086832005139782,
      "language_loss": 0.67982405,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70178866,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.647836446762085
    },
    {
      "auxiliary_loss_clip": 0.01144049,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.05107474,
      "balance_loss_mlp": 1.02026296,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 2.1913043554065754,
      "language_loss": 0.83955848,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.8612777,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.7013344764709473
    },
    {
      "auxiliary_loss_clip": 0.01171431,
      "auxiliary_loss_mlp": 0.0102381,
      "balance_loss_clip": 1.05592322,
      "balance_loss_mlp": 1.01597202,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 2.8224248293540115,
      "language_loss": 0.85848904,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88044143,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.706334352493286
    },
    {
      "auxiliary_loss_clip": 0.01169601,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.05407429,
      "balance_loss_mlp": 1.0189333,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.193535759719342,
      "language_loss": 0.70208216,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72405016,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.783169746398926
    },
    {
      "auxiliary_loss_clip": 0.01180195,
      "auxiliary_loss_mlp": 0.01022372,
      "balance_loss_clip": 1.05440521,
      "balance_loss_mlp": 1.01517797,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.891990854603247,
      "language_loss": 0.88464147,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90666711,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.6451613903045654
    },
    {
      "auxiliary_loss_clip": 0.01189721,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.05935907,
      "balance_loss_mlp": 1.02118301,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 3.3332447287806435,
      "language_loss": 0.86940658,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89160001,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 3.555262804031372
    },
    {
      "auxiliary_loss_clip": 0.01170791,
      "auxiliary_loss_mlp": 0.01024744,
      "balance_loss_clip": 1.05307937,
      "balance_loss_mlp": 1.01720405,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 6.537246930206938,
      "language_loss": 0.691001,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71295637,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.6257150173187256
    },
    {
      "auxiliary_loss_clip": 0.01156839,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.04752314,
      "balance_loss_mlp": 1.01643264,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.9486895712318182,
      "language_loss": 0.73224461,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75405359,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.7314350605010986
    },
    {
      "auxiliary_loss_clip": 0.01179261,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.05836201,
      "balance_loss_mlp": 1.01945233,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.798245372671371,
      "language_loss": 0.81902087,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84108669,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.5636298656463623
    },
    {
      "auxiliary_loss_clip": 0.01188873,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.06005049,
      "balance_loss_mlp": 1.01693535,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.0166416434288523,
      "language_loss": 0.87742823,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89956445,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 3.820436477661133
    },
    {
      "auxiliary_loss_clip": 0.01069821,
      "auxiliary_loss_mlp": 0.01002145,
      "balance_loss_clip": 1.0252378,
      "balance_loss_mlp": 1.00098884,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9874078014417073,
      "language_loss": 0.65039337,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67111301,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.1294264793395996
    },
    {
      "auxiliary_loss_clip": 0.01178744,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.05476546,
      "balance_loss_mlp": 1.02636254,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.477230071730182,
      "language_loss": 0.63731897,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65944898,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.628481388092041
    },
    {
      "auxiliary_loss_clip": 0.01183932,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.05691028,
      "balance_loss_mlp": 1.01803517,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 1.9311487616159573,
      "language_loss": 0.7990244,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.8211239,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.541278839111328
    },
    {
      "auxiliary_loss_clip": 0.01175965,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.05682862,
      "balance_loss_mlp": 1.01922393,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 1.9533111730274426,
      "language_loss": 0.80357528,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82560247,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.621910333633423
    },
    {
      "auxiliary_loss_clip": 0.01175569,
      "auxiliary_loss_mlp": 0.01026534,
      "balance_loss_clip": 1.0551604,
      "balance_loss_mlp": 1.01864839,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.571727880694069,
      "language_loss": 0.77542013,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79744118,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.6150333881378174
    },
    {
      "auxiliary_loss_clip": 0.01188697,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.05932999,
      "balance_loss_mlp": 1.02407432,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 1.9742290533740234,
      "language_loss": 0.78540206,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80761313,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 3.5998759269714355
    },
    {
      "auxiliary_loss_clip": 0.01093738,
      "auxiliary_loss_mlp": 0.01004937,
      "balance_loss_clip": 1.02657926,
      "balance_loss_mlp": 1.00397146,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8820858458746047,
      "language_loss": 0.6419034,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.6628902,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.9962069988250732
    },
    {
      "auxiliary_loss_clip": 0.01162451,
      "auxiliary_loss_mlp": 0.01024492,
      "balance_loss_clip": 1.05237365,
      "balance_loss_mlp": 1.01697016,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 9.122017883852362,
      "language_loss": 0.76437902,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78624845,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.675238609313965
    },
    {
      "auxiliary_loss_clip": 0.0114992,
      "auxiliary_loss_mlp": 0.01032107,
      "balance_loss_clip": 1.04880095,
      "balance_loss_mlp": 1.02444768,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 2.7541344171641198,
      "language_loss": 0.72483826,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74665856,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.726907730102539
    },
    {
      "auxiliary_loss_clip": 0.01081869,
      "auxiliary_loss_mlp": 0.01000945,
      "balance_loss_clip": 1.02494717,
      "balance_loss_mlp": 0.99992543,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1304780998026205,
      "language_loss": 0.55738419,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57821238,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.2313284873962402
    },
    {
      "auxiliary_loss_clip": 0.01150055,
      "auxiliary_loss_mlp": 0.01025105,
      "balance_loss_clip": 1.0490191,
      "balance_loss_mlp": 1.01773238,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.7816648720449053,
      "language_loss": 0.80793595,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.8296876,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.727376699447632
    },
    {
      "auxiliary_loss_clip": 0.01159427,
      "auxiliary_loss_mlp": 0.0102577,
      "balance_loss_clip": 1.05178714,
      "balance_loss_mlp": 1.0177052,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.3351124978319016,
      "language_loss": 0.74200618,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76385814,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.755647897720337
    },
    {
      "auxiliary_loss_clip": 0.0111746,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.04126406,
      "balance_loss_mlp": 1.02076519,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.7143544666796495,
      "language_loss": 0.79262453,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81408334,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.8429083824157715
    },
    {
      "auxiliary_loss_clip": 0.01157224,
      "auxiliary_loss_mlp": 0.00900737,
      "balance_loss_clip": 1.05094695,
      "balance_loss_mlp": 1.00066519,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 3.314087381577007,
      "language_loss": 0.82840204,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.84898168,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.6769332885742188
    },
    {
      "auxiliary_loss_clip": 0.0107104,
      "auxiliary_loss_mlp": 0.01003268,
      "balance_loss_clip": 1.02303433,
      "balance_loss_mlp": 1.00225449,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7304325163874835,
      "language_loss": 0.56063211,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58137524,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.2605247497558594
    },
    {
      "auxiliary_loss_clip": 0.01162638,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.05017853,
      "balance_loss_mlp": 1.0175215,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.0466460653521694,
      "language_loss": 0.71281427,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73469949,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 3.581159830093384
    },
    {
      "auxiliary_loss_clip": 0.01178955,
      "auxiliary_loss_mlp": 0.01020919,
      "balance_loss_clip": 1.05868006,
      "balance_loss_mlp": 1.01300955,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.1872303969711804,
      "language_loss": 0.8338393,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85583806,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.6177244186401367
    },
    {
      "auxiliary_loss_clip": 0.01181635,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.05599427,
      "balance_loss_mlp": 1.01918221,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.516529163735542,
      "language_loss": 0.71987045,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74194944,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.753232479095459
    },
    {
      "auxiliary_loss_clip": 0.01156573,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.05037344,
      "balance_loss_mlp": 1.02067173,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 1.7668428955927837,
      "language_loss": 0.80929446,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.83114409,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.7122857570648193
    },
    {
      "auxiliary_loss_clip": 0.0116443,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.05257034,
      "balance_loss_mlp": 1.02022791,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.795972766143387,
      "language_loss": 0.80217475,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.82409257,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.692436456680298
    },
    {
      "auxiliary_loss_clip": 0.01158595,
      "auxiliary_loss_mlp": 0.0102499,
      "balance_loss_clip": 1.05347741,
      "balance_loss_mlp": 1.01772404,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 5.2897101779651186,
      "language_loss": 0.69681168,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.71864754,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.700868844985962
    },
    {
      "auxiliary_loss_clip": 0.0116108,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.05378747,
      "balance_loss_mlp": 1.02154231,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.4872320912131283,
      "language_loss": 0.66660893,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68851483,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 3.682852029800415
    },
    {
      "auxiliary_loss_clip": 0.01184437,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.05539238,
      "balance_loss_mlp": 1.01807928,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.89278651535283,
      "language_loss": 0.78156096,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.80366039,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.663919687271118
    },
    {
      "auxiliary_loss_clip": 0.01151449,
      "auxiliary_loss_mlp": 0.00901248,
      "balance_loss_clip": 1.05299151,
      "balance_loss_mlp": 1.00059974,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.4620491913709937,
      "language_loss": 0.71939123,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.73991823,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.7980997562408447
    },
    {
      "auxiliary_loss_clip": 0.01160122,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.05139768,
      "balance_loss_mlp": 1.01961708,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 2.712765148744346,
      "language_loss": 0.78381741,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.80569214,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.6970489025115967
    },
    {
      "auxiliary_loss_clip": 0.01159217,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.05014992,
      "balance_loss_mlp": 1.02009237,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.084151746021029,
      "language_loss": 0.76255381,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78443152,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 3.6966445446014404
    },
    {
      "auxiliary_loss_clip": 0.01178577,
      "auxiliary_loss_mlp": 0.01032987,
      "balance_loss_clip": 1.05460405,
      "balance_loss_mlp": 1.02495241,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.302096748515497,
      "language_loss": 0.85442728,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87654293,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.6227004528045654
    },
    {
      "auxiliary_loss_clip": 0.01164732,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.04972112,
      "balance_loss_mlp": 1.02300584,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 2.001486809441405,
      "language_loss": 0.78955972,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81151921,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.702517509460449
    },
    {
      "auxiliary_loss_clip": 0.01175598,
      "auxiliary_loss_mlp": 0.0102265,
      "balance_loss_clip": 1.05315721,
      "balance_loss_mlp": 1.01496744,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.9624658878019343,
      "language_loss": 0.78346485,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80544734,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.6426944732666016
    },
    {
      "auxiliary_loss_clip": 0.01176216,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.05448854,
      "balance_loss_mlp": 1.02224374,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 1.9194096303788024,
      "language_loss": 0.85522658,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.87729257,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.5459249019622803
    },
    {
      "auxiliary_loss_clip": 0.01167934,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.0536443,
      "balance_loss_mlp": 1.0216372,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 1.763545400596143,
      "language_loss": 0.66051388,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68249017,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.909308433532715
    },
    {
      "auxiliary_loss_clip": 0.01136458,
      "auxiliary_loss_mlp": 0.01024594,
      "balance_loss_clip": 1.04426491,
      "balance_loss_mlp": 1.01555169,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 2.428163758914312,
      "language_loss": 0.79584312,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81745362,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 3.763639211654663
    },
    {
      "auxiliary_loss_clip": 0.0115921,
      "auxiliary_loss_mlp": 0.01030099,
      "balance_loss_clip": 1.04959488,
      "balance_loss_mlp": 1.02188826,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 2.162836633296888,
      "language_loss": 0.71446681,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73635995,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.637408971786499
    },
    {
      "auxiliary_loss_clip": 0.01166063,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.0509553,
      "balance_loss_mlp": 1.0199827,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 2.2071152290939366,
      "language_loss": 0.57558256,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59751689,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.6968905925750732
    },
    {
      "auxiliary_loss_clip": 0.01166913,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.05445862,
      "balance_loss_mlp": 1.01849198,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.5537906522220106,
      "language_loss": 0.77121627,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79314631,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.6468729972839355
    },
    {
      "auxiliary_loss_clip": 0.01183885,
      "auxiliary_loss_mlp": 0.01028623,
      "balance_loss_clip": 1.05716896,
      "balance_loss_mlp": 1.02063024,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.9173907995443586,
      "language_loss": 0.67380315,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69592822,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.578444719314575
    },
    {
      "auxiliary_loss_clip": 0.01181151,
      "auxiliary_loss_mlp": 0.01022591,
      "balance_loss_clip": 1.05401015,
      "balance_loss_mlp": 1.01508069,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 2.053530215612958,
      "language_loss": 0.80359006,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82562745,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.5915608406066895
    },
    {
      "auxiliary_loss_clip": 0.01174692,
      "auxiliary_loss_mlp": 0.00900457,
      "balance_loss_clip": 1.05575442,
      "balance_loss_mlp": 1.0005908,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.9733083920650427,
      "language_loss": 0.84296501,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86371648,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.6859662532806396
    },
    {
      "auxiliary_loss_clip": 0.01172545,
      "auxiliary_loss_mlp": 0.0102931,
      "balance_loss_clip": 1.05653739,
      "balance_loss_mlp": 1.02095985,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 1.9869195766301895,
      "language_loss": 0.80273592,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82475448,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.6721880435943604
    },
    {
      "auxiliary_loss_clip": 0.0108196,
      "auxiliary_loss_mlp": 0.01001009,
      "balance_loss_clip": 1.02526987,
      "balance_loss_mlp": 0.99997753,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.8305764105657669,
      "language_loss": 0.56651855,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58734822,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.372314929962158
    },
    {
      "auxiliary_loss_clip": 0.01176523,
      "auxiliary_loss_mlp": 0.01022156,
      "balance_loss_clip": 1.05666292,
      "balance_loss_mlp": 1.01382947,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.7637748890461773,
      "language_loss": 0.77421105,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79619777,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.6092915534973145
    },
    {
      "auxiliary_loss_clip": 0.01182189,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.05446863,
      "balance_loss_mlp": 1.01637673,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.706770836605696,
      "language_loss": 0.80510056,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.8271662,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 3.523588180541992
    },
    {
      "auxiliary_loss_clip": 0.01156749,
      "auxiliary_loss_mlp": 0.0102225,
      "balance_loss_clip": 1.04974627,
      "balance_loss_mlp": 1.01458502,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 1.9360589322990518,
      "language_loss": 0.79867947,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.8204695,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.6614744663238525
    },
    {
      "auxiliary_loss_clip": 0.01176084,
      "auxiliary_loss_mlp": 0.01023719,
      "balance_loss_clip": 1.05374408,
      "balance_loss_mlp": 1.01654243,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 2.1909003742796194,
      "language_loss": 0.77197719,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.79397517,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.601841449737549
    },
    {
      "auxiliary_loss_clip": 0.01165696,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.051548,
      "balance_loss_mlp": 1.01625562,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.735917330072044,
      "language_loss": 0.80672467,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.82861984,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.6583869457244873
    },
    {
      "auxiliary_loss_clip": 0.01163508,
      "auxiliary_loss_mlp": 0.01027842,
      "balance_loss_clip": 1.05073571,
      "balance_loss_mlp": 1.01996207,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.965418343000083,
      "language_loss": 0.89757276,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.91948628,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.5901145935058594
    },
    {
      "auxiliary_loss_clip": 0.01148753,
      "auxiliary_loss_mlp": 0.01024799,
      "balance_loss_clip": 1.04571891,
      "balance_loss_mlp": 1.01709771,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.926260281412229,
      "language_loss": 0.73396748,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75570297,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.7878310680389404
    },
    {
      "auxiliary_loss_clip": 0.01188252,
      "auxiliary_loss_mlp": 0.01027889,
      "balance_loss_clip": 1.05916381,
      "balance_loss_mlp": 1.01925206,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.661534924671136,
      "language_loss": 0.6940915,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71625292,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.7935919761657715
    },
    {
      "auxiliary_loss_clip": 0.01180125,
      "auxiliary_loss_mlp": 0.01030983,
      "balance_loss_clip": 1.05624008,
      "balance_loss_mlp": 1.02271557,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 5.0671339746666275,
      "language_loss": 0.6931963,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71530741,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 3.5764477252960205
    },
    {
      "auxiliary_loss_clip": 0.01183286,
      "auxiliary_loss_mlp": 0.00901383,
      "balance_loss_clip": 1.05547857,
      "balance_loss_mlp": 1.00067556,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.536423549707332,
      "language_loss": 0.84649515,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.86734188,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.628394842147827
    },
    {
      "auxiliary_loss_clip": 0.01080699,
      "auxiliary_loss_mlp": 0.01002511,
      "balance_loss_clip": 1.02680302,
      "balance_loss_mlp": 1.0014739,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7263373091638778,
      "language_loss": 0.56896251,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.58979464,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.1397924423217773
    },
    {
      "auxiliary_loss_clip": 0.01175062,
      "auxiliary_loss_mlp": 0.00901511,
      "balance_loss_clip": 1.05626559,
      "balance_loss_mlp": 1.00074756,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.192857239056569,
      "language_loss": 0.76261175,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.78337747,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 3.5726139545440674
    },
    {
      "auxiliary_loss_clip": 0.01165423,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.05136633,
      "balance_loss_mlp": 1.02187753,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.127451118869225,
      "language_loss": 0.77409494,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79604495,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.681370496749878
    },
    {
      "auxiliary_loss_clip": 0.01163131,
      "auxiliary_loss_mlp": 0.01023604,
      "balance_loss_clip": 1.05356598,
      "balance_loss_mlp": 1.01615024,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.448281767994828,
      "language_loss": 0.82562053,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84748793,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.6584134101867676
    },
    {
      "auxiliary_loss_clip": 0.01171155,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.05328727,
      "balance_loss_mlp": 1.01848984,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.7596636778420793,
      "language_loss": 0.81517684,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83714753,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.6972262859344482
    },
    {
      "auxiliary_loss_clip": 0.01142622,
      "auxiliary_loss_mlp": 0.01025752,
      "balance_loss_clip": 1.04902124,
      "balance_loss_mlp": 1.01765156,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.6441308311993854,
      "language_loss": 0.83228487,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85396862,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.7585716247558594
    },
    {
      "auxiliary_loss_clip": 0.01178774,
      "auxiliary_loss_mlp": 0.00901295,
      "balance_loss_clip": 1.05541945,
      "balance_loss_mlp": 1.00062871,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.304178054746385,
      "language_loss": 0.66576302,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.68656373,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.6974596977233887
    },
    {
      "auxiliary_loss_clip": 0.01173187,
      "auxiliary_loss_mlp": 0.01036325,
      "balance_loss_clip": 1.05295587,
      "balance_loss_mlp": 1.027825,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 1.9965132508689376,
      "language_loss": 0.80787623,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.82997131,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.6656622886657715
    },
    {
      "auxiliary_loss_clip": 0.01178867,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.05644095,
      "balance_loss_mlp": 1.0201087,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 3.246831883324828,
      "language_loss": 0.72430086,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74636602,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 3.5494117736816406
    },
    {
      "auxiliary_loss_clip": 0.01146413,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.05137455,
      "balance_loss_mlp": 1.02168822,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 1.7853847777543888,
      "language_loss": 0.81040382,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83216298,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.6816415786743164
    },
    {
      "auxiliary_loss_clip": 0.01182795,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.05549598,
      "balance_loss_mlp": 1.02104342,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.638268843301154,
      "language_loss": 0.81849819,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84061748,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.6252918243408203
    },
    {
      "auxiliary_loss_clip": 0.01172953,
      "auxiliary_loss_mlp": 0.01023838,
      "balance_loss_clip": 1.05539572,
      "balance_loss_mlp": 1.01649213,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 4.60012758514203,
      "language_loss": 0.80156386,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.82353181,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.6507067680358887
    },
    {
      "auxiliary_loss_clip": 0.01175773,
      "auxiliary_loss_mlp": 0.01029648,
      "balance_loss_clip": 1.05385149,
      "balance_loss_mlp": 1.02173841,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 1.9857021663913093,
      "language_loss": 0.71167886,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73373306,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.5870580673217773
    },
    {
      "auxiliary_loss_clip": 0.01179221,
      "auxiliary_loss_mlp": 0.01033309,
      "balance_loss_clip": 1.05350709,
      "balance_loss_mlp": 1.02440464,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 3.787273969151755,
      "language_loss": 0.80300021,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82512558,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.717883825302124
    },
    {
      "auxiliary_loss_clip": 0.01172269,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.05167496,
      "balance_loss_mlp": 1.01725507,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.8443601251746211,
      "language_loss": 0.85806131,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.88003707,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.6567537784576416
    },
    {
      "auxiliary_loss_clip": 0.0115118,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.05086553,
      "balance_loss_mlp": 1.01943171,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 2.561614778558238,
      "language_loss": 0.66631997,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68811178,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.739225387573242
    },
    {
      "auxiliary_loss_clip": 0.0116886,
      "auxiliary_loss_mlp": 0.01028077,
      "balance_loss_clip": 1.05252337,
      "balance_loss_mlp": 1.02033424,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.4453953853980348,
      "language_loss": 0.77503586,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.7970053,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.7417874336242676
    },
    {
      "auxiliary_loss_clip": 0.01168411,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.05399132,
      "balance_loss_mlp": 1.01887763,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 1.8709156719642481,
      "language_loss": 0.57535923,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.5973137,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.7161717414855957
    },
    {
      "auxiliary_loss_clip": 0.01161736,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.05304539,
      "balance_loss_mlp": 1.01699162,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 1.980644259337521,
      "language_loss": 0.83003247,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.85189605,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.6090457439422607
    },
    {
      "auxiliary_loss_clip": 0.01165794,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.05259824,
      "balance_loss_mlp": 1.02097201,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.281676369422857,
      "language_loss": 0.77812183,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.8000679,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.704953670501709
    },
    {
      "auxiliary_loss_clip": 0.01156745,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.05246985,
      "balance_loss_mlp": 1.02096188,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.5915979787673242,
      "language_loss": 0.7441783,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76603162,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.717211961746216
    },
    {
      "auxiliary_loss_clip": 0.0117524,
      "auxiliary_loss_mlp": 0.01029603,
      "balance_loss_clip": 1.05589044,
      "balance_loss_mlp": 1.02198589,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 1.9120110052006165,
      "language_loss": 0.89889961,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.92094803,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.6261911392211914
    },
    {
      "auxiliary_loss_clip": 0.01161418,
      "auxiliary_loss_mlp": 0.01028237,
      "balance_loss_clip": 1.05113101,
      "balance_loss_mlp": 1.01973736,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 1.8576786525651763,
      "language_loss": 0.63491791,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.65681446,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.744452953338623
    },
    {
      "auxiliary_loss_clip": 0.01167744,
      "auxiliary_loss_mlp": 0.01032346,
      "balance_loss_clip": 1.05450356,
      "balance_loss_mlp": 1.02378631,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 2.2450174295907748,
      "language_loss": 0.84597641,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86797726,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.6547787189483643
    },
    {
      "auxiliary_loss_clip": 0.01181612,
      "auxiliary_loss_mlp": 0.0090075,
      "balance_loss_clip": 1.05692112,
      "balance_loss_mlp": 1.00074148,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.759733226850083,
      "language_loss": 0.76616848,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78699207,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.6710057258605957
    },
    {
      "auxiliary_loss_clip": 0.01178069,
      "auxiliary_loss_mlp": 0.0090065,
      "balance_loss_clip": 1.05744207,
      "balance_loss_mlp": 1.00079024,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 2.195056158075301,
      "language_loss": 0.81191933,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.83270651,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 3.6416332721710205
    },
    {
      "auxiliary_loss_clip": 0.01176932,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.05801868,
      "balance_loss_mlp": 1.0190928,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.150595957282289,
      "language_loss": 0.72795069,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.74999011,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.5971474647521973
    },
    {
      "auxiliary_loss_clip": 0.01139332,
      "auxiliary_loss_mlp": 0.00900854,
      "balance_loss_clip": 1.0506711,
      "balance_loss_mlp": 1.00074422,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 2.380134111820999,
      "language_loss": 0.73331785,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75371969,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.769850969314575
    },
    {
      "auxiliary_loss_clip": 0.01183045,
      "auxiliary_loss_mlp": 0.00901804,
      "balance_loss_clip": 1.05549145,
      "balance_loss_mlp": 1.00075996,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 2.027257350135159,
      "language_loss": 0.68091118,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.70175964,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 3.4023663997650146
    },
    {
      "auxiliary_loss_clip": 0.0114625,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.04977322,
      "balance_loss_mlp": 1.02189755,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 2.172148250865477,
      "language_loss": 0.8786369,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.90039718,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.7710213661193848
    },
    {
      "auxiliary_loss_clip": 0.0109156,
      "auxiliary_loss_mlp": 0.00891429,
      "balance_loss_clip": 1.02621222,
      "balance_loss_mlp": 0.99991339,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7843119941547215,
      "language_loss": 0.61766744,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63749731,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.297001838684082
    },
    {
      "auxiliary_loss_clip": 0.01156676,
      "auxiliary_loss_mlp": 0.01023029,
      "balance_loss_clip": 1.05273747,
      "balance_loss_mlp": 1.01513743,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 2.0519612265230243,
      "language_loss": 0.78308117,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80487823,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.707089424133301
    },
    {
      "auxiliary_loss_clip": 0.01167833,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.05207396,
      "balance_loss_mlp": 1.01990712,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.0629251028917444,
      "language_loss": 0.73898947,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.76094562,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.7842557430267334
    },
    {
      "auxiliary_loss_clip": 0.01183249,
      "auxiliary_loss_mlp": 0.01028555,
      "balance_loss_clip": 1.05599403,
      "balance_loss_mlp": 1.0204134,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 3.186890714185193,
      "language_loss": 0.74427688,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76639485,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.697317123413086
    },
    {
      "auxiliary_loss_clip": 0.01173805,
      "auxiliary_loss_mlp": 0.01024301,
      "balance_loss_clip": 1.05365574,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.4535694196325624,
      "language_loss": 0.67789984,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.69988096,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 3.6324896812438965
    },
    {
      "auxiliary_loss_clip": 0.01155148,
      "auxiliary_loss_mlp": 0.0103405,
      "balance_loss_clip": 1.05640137,
      "balance_loss_mlp": 1.02642024,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.5332669725325285,
      "language_loss": 0.73459554,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75648749,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.7201175689697266
    },
    {
      "auxiliary_loss_clip": 0.01176676,
      "auxiliary_loss_mlp": 0.01027578,
      "balance_loss_clip": 1.05510294,
      "balance_loss_mlp": 1.01896477,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 2.0173221839744784,
      "language_loss": 0.79150224,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81354481,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.6426541805267334
    },
    {
      "auxiliary_loss_clip": 0.01088161,
      "auxiliary_loss_mlp": 0.01001245,
      "balance_loss_clip": 1.02598739,
      "balance_loss_mlp": 1.0001955,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.770277981583272,
      "language_loss": 0.59523857,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61613262,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.2050890922546387
    },
    {
      "auxiliary_loss_clip": 0.01177454,
      "auxiliary_loss_mlp": 0.01027078,
      "balance_loss_clip": 1.05732715,
      "balance_loss_mlp": 1.01910031,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 1.8820330231713451,
      "language_loss": 0.71805477,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74010015,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.5805251598358154
    },
    {
      "auxiliary_loss_clip": 0.01166299,
      "auxiliary_loss_mlp": 0.0102397,
      "balance_loss_clip": 1.05345285,
      "balance_loss_mlp": 1.01583982,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 2.267083907525934,
      "language_loss": 0.62741435,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.64931697,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.6856696605682373
    },
    {
      "auxiliary_loss_clip": 0.01169649,
      "auxiliary_loss_mlp": 0.0102974,
      "balance_loss_clip": 1.0547756,
      "balance_loss_mlp": 1.02191961,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.212075289613059,
      "language_loss": 0.85537136,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87736529,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.6041955947875977
    },
    {
      "auxiliary_loss_clip": 0.01186601,
      "auxiliary_loss_mlp": 0.01026894,
      "balance_loss_clip": 1.05724227,
      "balance_loss_mlp": 1.01875782,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.2277893267618403,
      "language_loss": 0.79451907,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81665409,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.6830220222473145
    },
    {
      "auxiliary_loss_clip": 0.01167344,
      "auxiliary_loss_mlp": 0.01030036,
      "balance_loss_clip": 1.05713534,
      "balance_loss_mlp": 1.02177501,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 3.420556606252503,
      "language_loss": 0.7160939,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73806775,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.675870895385742
    },
    {
      "auxiliary_loss_clip": 0.01183749,
      "auxiliary_loss_mlp": 0.01027285,
      "balance_loss_clip": 1.05769312,
      "balance_loss_mlp": 1.0188098,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.7631695238066794,
      "language_loss": 0.67895091,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.70106125,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.7721076011657715
    },
    {
      "auxiliary_loss_clip": 0.0116592,
      "auxiliary_loss_mlp": 0.01028372,
      "balance_loss_clip": 1.05437791,
      "balance_loss_mlp": 1.02010441,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 2.264066933332072,
      "language_loss": 0.73677403,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75871694,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 3.6353251934051514
    },
    {
      "auxiliary_loss_clip": 0.01160439,
      "auxiliary_loss_mlp": 0.01031718,
      "balance_loss_clip": 1.05491042,
      "balance_loss_mlp": 1.02426124,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.6340222254805232,
      "language_loss": 0.69600785,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.71792948,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.754420042037964
    },
    {
      "auxiliary_loss_clip": 0.01174716,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.05116761,
      "balance_loss_mlp": 1.01761699,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.929012224655766,
      "language_loss": 0.86069834,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88270152,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.659552812576294
    },
    {
      "auxiliary_loss_clip": 0.01159831,
      "auxiliary_loss_mlp": 0.01028233,
      "balance_loss_clip": 1.0511657,
      "balance_loss_mlp": 1.02003169,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.0909052958544483,
      "language_loss": 0.86240828,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88428897,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.6144938468933105
    },
    {
      "auxiliary_loss_clip": 0.01160151,
      "auxiliary_loss_mlp": 0.01026828,
      "balance_loss_clip": 1.05645323,
      "balance_loss_mlp": 1.01892424,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 2.2623202353868663,
      "language_loss": 0.7602405,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78211021,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.699049949645996
    },
    {
      "auxiliary_loss_clip": 0.01158695,
      "auxiliary_loss_mlp": 0.01028953,
      "balance_loss_clip": 1.05385971,
      "balance_loss_mlp": 1.02119303,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.7707077099242101,
      "language_loss": 0.83989471,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.86177123,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 2.6561617851257324
    },
    {
      "auxiliary_loss_clip": 0.01178214,
      "auxiliary_loss_mlp": 0.00901932,
      "balance_loss_clip": 1.05842793,
      "balance_loss_mlp": 1.00075448,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.5969578948245586,
      "language_loss": 0.75567234,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77647376,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.6750364303588867
    },
    {
      "auxiliary_loss_clip": 0.0117469,
      "auxiliary_loss_mlp": 0.01025191,
      "balance_loss_clip": 1.05645251,
      "balance_loss_mlp": 1.01747787,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.8527659264374006,
      "language_loss": 0.83147693,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85347569,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 3.555737018585205
    },
    {
      "auxiliary_loss_clip": 0.01160449,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.05007589,
      "balance_loss_mlp": 1.01740623,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.9667697275318685,
      "language_loss": 0.77354854,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79540741,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.633460760116577
    },
    {
      "auxiliary_loss_clip": 0.0116118,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.04862094,
      "balance_loss_mlp": 1.01894605,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.171011942836368,
      "language_loss": 0.80731088,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82919598,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.7070794105529785
    },
    {
      "auxiliary_loss_clip": 0.01165284,
      "auxiliary_loss_mlp": 0.0090137,
      "balance_loss_clip": 1.05382538,
      "balance_loss_mlp": 1.00065136,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.009767347992307,
      "language_loss": 0.66413498,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68480152,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 3.6937320232391357
    },
    {
      "auxiliary_loss_clip": 0.01182866,
      "auxiliary_loss_mlp": 0.01026368,
      "balance_loss_clip": 1.05719328,
      "balance_loss_mlp": 1.01804757,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 3.786193303950328,
      "language_loss": 0.85822701,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88031936,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.675224781036377
    },
    {
      "auxiliary_loss_clip": 0.01183402,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.05527234,
      "balance_loss_mlp": 1.02042365,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.8190402116929985,
      "language_loss": 0.68515372,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70727831,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.6849069595336914
    },
    {
      "auxiliary_loss_clip": 0.01168003,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.05691683,
      "balance_loss_mlp": 1.01818919,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 4.119366481655786,
      "language_loss": 0.74372756,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76566851,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.75541090965271
    },
    {
      "auxiliary_loss_clip": 0.01127062,
      "auxiliary_loss_mlp": 0.0102355,
      "balance_loss_clip": 1.04702401,
      "balance_loss_mlp": 1.01587915,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 2.192798259583894,
      "language_loss": 0.77280462,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79431069,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.73004412651062
    },
    {
      "auxiliary_loss_clip": 0.01176286,
      "auxiliary_loss_mlp": 0.01029763,
      "balance_loss_clip": 1.05474854,
      "balance_loss_mlp": 1.0215199,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.3621918773376884,
      "language_loss": 0.71080875,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.73286921,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.658717632293701
    },
    {
      "auxiliary_loss_clip": 0.01178617,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 1.05541301,
      "balance_loss_mlp": 1.01888108,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 2.030681424485761,
      "language_loss": 0.8688271,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89087683,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.640010356903076
    },
    {
      "auxiliary_loss_clip": 0.01175181,
      "auxiliary_loss_mlp": 0.00902133,
      "balance_loss_clip": 1.05428958,
      "balance_loss_mlp": 1.00079823,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 2.568485307820161,
      "language_loss": 0.64403993,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66481304,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 3.639141082763672
    },
    {
      "auxiliary_loss_clip": 0.01160596,
      "auxiliary_loss_mlp": 0.01030273,
      "balance_loss_clip": 1.0519675,
      "balance_loss_mlp": 1.02213049,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 6.86992122079857,
      "language_loss": 0.90538043,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92728913,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.778740167617798
    },
    {
      "auxiliary_loss_clip": 0.01164678,
      "auxiliary_loss_mlp": 0.01026173,
      "balance_loss_clip": 1.05388117,
      "balance_loss_mlp": 1.01828158,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.7749981881996366,
      "language_loss": 0.6460166,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66792512,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.650973081588745
    },
    {
      "auxiliary_loss_clip": 0.01163282,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.05222178,
      "balance_loss_mlp": 1.01748657,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.892762870935997,
      "language_loss": 0.70941985,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.73131156,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.6259703636169434
    },
    {
      "auxiliary_loss_clip": 0.01164762,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.05092835,
      "balance_loss_mlp": 1.02020085,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 3.286131965960251,
      "language_loss": 0.8088429,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.83077073,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.629316806793213
    },
    {
      "auxiliary_loss_clip": 0.01169444,
      "auxiliary_loss_mlp": 0.01034611,
      "balance_loss_clip": 1.05651772,
      "balance_loss_mlp": 1.02685106,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.996544235322276,
      "language_loss": 0.71854305,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74058366,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.7046780586242676
    },
    {
      "auxiliary_loss_clip": 0.0115148,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05271137,
      "balance_loss_mlp": 1.021137,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.6334318902744793,
      "language_loss": 0.72809267,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74990797,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.8043107986450195
    },
    {
      "auxiliary_loss_clip": 0.0116081,
      "auxiliary_loss_mlp": 0.01028185,
      "balance_loss_clip": 1.04832602,
      "balance_loss_mlp": 1.02066326,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.2475802013786295,
      "language_loss": 0.72448903,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.7463789,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.6736135482788086
    },
    {
      "auxiliary_loss_clip": 0.01182671,
      "auxiliary_loss_mlp": 0.01029891,
      "balance_loss_clip": 1.05541754,
      "balance_loss_mlp": 1.02133214,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.9173082371938563,
      "language_loss": 0.80078006,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82290572,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.618041515350342
    },
    {
      "auxiliary_loss_clip": 0.0117272,
      "auxiliary_loss_mlp": 0.01027967,
      "balance_loss_clip": 1.05622053,
      "balance_loss_mlp": 1.01947367,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.142705190026529,
      "language_loss": 0.84085441,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.86286122,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.6891026496887207
    },
    {
      "auxiliary_loss_clip": 0.01174243,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.05733633,
      "balance_loss_mlp": 1.01691151,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.7629634169136206,
      "language_loss": 0.74241078,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76439917,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 3.6165337562561035
    },
    {
      "auxiliary_loss_clip": 0.0113361,
      "auxiliary_loss_mlp": 0.01027534,
      "balance_loss_clip": 1.04665816,
      "balance_loss_mlp": 1.0196538,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.8093066694312343,
      "language_loss": 0.76153588,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78314734,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.717043399810791
    },
    {
      "auxiliary_loss_clip": 0.01185923,
      "auxiliary_loss_mlp": 0.01023586,
      "balance_loss_clip": 1.05865932,
      "balance_loss_mlp": 1.01620126,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 5.907167430449834,
      "language_loss": 0.76218474,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78427982,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.6079273223876953
    },
    {
      "auxiliary_loss_clip": 0.01074024,
      "auxiliary_loss_mlp": 0.01001846,
      "balance_loss_clip": 1.02406752,
      "balance_loss_mlp": 1.00064838,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7576262441471925,
      "language_loss": 0.56499237,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58575106,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.2429473400115967
    },
    {
      "auxiliary_loss_clip": 0.01186842,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 1.05734289,
      "balance_loss_mlp": 1.01904666,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.1773389981905087,
      "language_loss": 0.89578092,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91792184,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 2.5440187454223633
    },
    {
      "auxiliary_loss_clip": 0.01093756,
      "auxiliary_loss_mlp": 0.01001057,
      "balance_loss_clip": 1.0269531,
      "balance_loss_mlp": 0.99989432,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 1.038499139636523,
      "language_loss": 0.56128603,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58223414,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.1711134910583496
    },
    {
      "auxiliary_loss_clip": 0.01174964,
      "auxiliary_loss_mlp": 0.0102595,
      "balance_loss_clip": 1.05717492,
      "balance_loss_mlp": 1.01775444,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 2.392720285938847,
      "language_loss": 0.80719721,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.82920635,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 3.5692355632781982
    },
    {
      "auxiliary_loss_clip": 0.0117953,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.05679512,
      "balance_loss_mlp": 1.01912665,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.8952255262774726,
      "language_loss": 0.76323867,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78530848,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.731264591217041
    },
    {
      "auxiliary_loss_clip": 0.01182243,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.05598438,
      "balance_loss_mlp": 1.01888943,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 2.3104937987009464,
      "language_loss": 0.82676595,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84885865,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.5456016063690186
    },
    {
      "auxiliary_loss_clip": 0.01171223,
      "auxiliary_loss_mlp": 0.00901149,
      "balance_loss_clip": 1.05465961,
      "balance_loss_mlp": 1.00101328,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 2.325037941834699,
      "language_loss": 0.70995045,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.73067415,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.6712846755981445
    },
    {
      "auxiliary_loss_clip": 0.01135811,
      "auxiliary_loss_mlp": 0.01030001,
      "balance_loss_clip": 1.04680979,
      "balance_loss_mlp": 1.02137649,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 2.3007803220480234,
      "language_loss": 0.75326413,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77492225,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 3.7007195949554443
    },
    {
      "auxiliary_loss_clip": 0.01165603,
      "auxiliary_loss_mlp": 0.01023121,
      "balance_loss_clip": 1.05483937,
      "balance_loss_mlp": 1.01567602,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.8363167993163454,
      "language_loss": 0.7651633,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78705049,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.6238365173339844
    },
    {
      "auxiliary_loss_clip": 0.01145122,
      "auxiliary_loss_mlp": 0.01022533,
      "balance_loss_clip": 1.05212057,
      "balance_loss_mlp": 1.01493943,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.684153304734278,
      "language_loss": 0.72171682,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74339342,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.80094313621521
    },
    {
      "auxiliary_loss_clip": 0.01175817,
      "auxiliary_loss_mlp": 0.01030582,
      "balance_loss_clip": 1.0578469,
      "balance_loss_mlp": 1.02220738,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.464978303865813,
      "language_loss": 0.71894562,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74100953,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.617182970046997
    },
    {
      "auxiliary_loss_clip": 0.01187025,
      "auxiliary_loss_mlp": 0.01032926,
      "balance_loss_clip": 1.05757141,
      "balance_loss_mlp": 1.02441478,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.6069653849915568,
      "language_loss": 0.78563714,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.80783665,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.755795955657959
    },
    {
      "auxiliary_loss_clip": 0.01163993,
      "auxiliary_loss_mlp": 0.01031201,
      "balance_loss_clip": 1.05556178,
      "balance_loss_mlp": 1.02299905,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.1452666159395326,
      "language_loss": 0.73992717,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.76187921,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.692185640335083
    },
    {
      "auxiliary_loss_clip": 0.01155121,
      "auxiliary_loss_mlp": 0.01025185,
      "balance_loss_clip": 1.05037248,
      "balance_loss_mlp": 1.01700473,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 2.015629420064419,
      "language_loss": 0.85238725,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87419033,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 3.7966690063476562
    },
    {
      "auxiliary_loss_clip": 0.01177378,
      "auxiliary_loss_mlp": 0.01032187,
      "balance_loss_clip": 1.05380726,
      "balance_loss_mlp": 1.02391386,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.7697987832142252,
      "language_loss": 0.71012652,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.7322222,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.6121509075164795
    },
    {
      "auxiliary_loss_clip": 0.01150423,
      "auxiliary_loss_mlp": 0.01031365,
      "balance_loss_clip": 1.05399144,
      "balance_loss_mlp": 1.0236398,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 2.1130562629511553,
      "language_loss": 0.75028247,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77210033,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.7197375297546387
    },
    {
      "auxiliary_loss_clip": 0.01161232,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.05542159,
      "balance_loss_mlp": 1.01953602,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 2.631167885261964,
      "language_loss": 0.77347469,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79535675,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.6972098350524902
    },
    {
      "auxiliary_loss_clip": 0.01177637,
      "auxiliary_loss_mlp": 0.0102698,
      "balance_loss_clip": 1.05772161,
      "balance_loss_mlp": 1.01857018,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.930085005380291,
      "language_loss": 0.81418514,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83623135,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.65942120552063
    },
    {
      "auxiliary_loss_clip": 0.01163909,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.05377674,
      "balance_loss_mlp": 1.02124143,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.4789410820369944,
      "language_loss": 0.84064627,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86258137,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.7761764526367188
    },
    {
      "auxiliary_loss_clip": 0.01175401,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.05731976,
      "balance_loss_mlp": 1.01618981,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 3.2660994705005417,
      "language_loss": 0.87272179,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.89471734,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.621668815612793
    },
    {
      "auxiliary_loss_clip": 0.01180326,
      "auxiliary_loss_mlp": 0.01032855,
      "balance_loss_clip": 1.05598211,
      "balance_loss_mlp": 1.02434969,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 2.4128836430535374,
      "language_loss": 0.7374084,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.7595402,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.6820995807647705
    },
    {
      "auxiliary_loss_clip": 0.0118027,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.05396891,
      "balance_loss_mlp": 1.01912498,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 1.9594062703812167,
      "language_loss": 0.60138673,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62345618,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.5651237964630127
    },
    {
      "auxiliary_loss_clip": 0.01154211,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 1.04989171,
      "balance_loss_mlp": 1.02008271,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.64470352425171,
      "language_loss": 0.68861783,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71044499,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.7451584339141846
    },
    {
      "auxiliary_loss_clip": 0.01171324,
      "auxiliary_loss_mlp": 0.01028213,
      "balance_loss_clip": 1.05738187,
      "balance_loss_mlp": 1.020679,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 1.9655508372108554,
      "language_loss": 0.69341379,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71540916,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 3.58382511138916
    },
    {
      "auxiliary_loss_clip": 0.01154221,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04743814,
      "balance_loss_mlp": 1.01706314,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 4.9485969413443565,
      "language_loss": 0.74321437,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76501042,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.7314460277557373
    },
    {
      "auxiliary_loss_clip": 0.01166785,
      "auxiliary_loss_mlp": 0.01028467,
      "balance_loss_clip": 1.05199766,
      "balance_loss_mlp": 1.01968741,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 2.0199534412742137,
      "language_loss": 0.72995448,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75190699,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.7302544116973877
    },
    {
      "auxiliary_loss_clip": 0.01153337,
      "auxiliary_loss_mlp": 0.01027297,
      "balance_loss_clip": 1.0482955,
      "balance_loss_mlp": 1.01936364,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 2.316119217012114,
      "language_loss": 0.63464916,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65645546,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.727280855178833
    },
    {
      "auxiliary_loss_clip": 0.01067939,
      "auxiliary_loss_mlp": 0.01001578,
      "balance_loss_clip": 1.02185345,
      "balance_loss_mlp": 1.00051677,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8076458578600028,
      "language_loss": 0.55811918,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57881427,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.2011682987213135
    },
    {
      "auxiliary_loss_clip": 0.01160769,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 1.05391717,
      "balance_loss_mlp": 1.01672792,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.3343268603371237,
      "language_loss": 0.67139578,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.69323719,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.6858184337615967
    },
    {
      "auxiliary_loss_clip": 0.01182325,
      "auxiliary_loss_mlp": 0.00901276,
      "balance_loss_clip": 1.05640173,
      "balance_loss_mlp": 1.00094104,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.845420336080643,
      "language_loss": 0.81141806,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.83225405,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.6814804077148438
    },
    {
      "auxiliary_loss_clip": 0.01160843,
      "auxiliary_loss_mlp": 0.00901124,
      "balance_loss_clip": 1.05521035,
      "balance_loss_mlp": 1.00097799,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.973094625997785,
      "language_loss": 0.82184887,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84246862,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 3.644103527069092
    },
    {
      "auxiliary_loss_clip": 0.01179336,
      "auxiliary_loss_mlp": 0.01034587,
      "balance_loss_clip": 1.05480599,
      "balance_loss_mlp": 1.02636766,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.6884843679963752,
      "language_loss": 0.8264522,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84859145,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.6938984394073486
    },
    {
      "auxiliary_loss_clip": 0.01168043,
      "auxiliary_loss_mlp": 0.01024866,
      "balance_loss_clip": 1.05322778,
      "balance_loss_mlp": 1.01716554,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.620389633744288,
      "language_loss": 0.86004281,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.8819719,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.7088935375213623
    },
    {
      "auxiliary_loss_clip": 0.01170487,
      "auxiliary_loss_mlp": 0.01026365,
      "balance_loss_clip": 1.05363119,
      "balance_loss_mlp": 1.01841998,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 2.1136844923231304,
      "language_loss": 0.72945529,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75142384,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 3.565429449081421
    },
    {
      "auxiliary_loss_clip": 0.01108063,
      "auxiliary_loss_mlp": 0.01036732,
      "balance_loss_clip": 1.0424726,
      "balance_loss_mlp": 1.02909112,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 2.384883689721465,
      "language_loss": 0.80206418,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82351208,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 3.0590627193450928
    },
    {
      "auxiliary_loss_clip": 0.01164859,
      "auxiliary_loss_mlp": 0.01029327,
      "balance_loss_clip": 1.05957484,
      "balance_loss_mlp": 1.02133369,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 2.285681969559509,
      "language_loss": 0.77794784,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.79988968,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 3.045557975769043
    },
    {
      "auxiliary_loss_clip": 0.01160833,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.05210721,
      "balance_loss_mlp": 1.01958656,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.4682372888634303,
      "language_loss": 0.80507958,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.8269645,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.6061208248138428
    },
    {
      "auxiliary_loss_clip": 0.01169573,
      "auxiliary_loss_mlp": 0.01030863,
      "balance_loss_clip": 1.05255878,
      "balance_loss_mlp": 1.02310252,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 1.9302953222960926,
      "language_loss": 0.82478195,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84678632,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.623866319656372
    },
    {
      "auxiliary_loss_clip": 0.01122908,
      "auxiliary_loss_mlp": 0.01025788,
      "balance_loss_clip": 1.04421425,
      "balance_loss_mlp": 1.01771748,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.6608570469392125,
      "language_loss": 0.77429402,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79578102,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.758359670639038
    },
    {
      "auxiliary_loss_clip": 0.01131037,
      "auxiliary_loss_mlp": 0.01024847,
      "balance_loss_clip": 1.0423193,
      "balance_loss_mlp": 1.01683021,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 1.65419950384643,
      "language_loss": 0.79748809,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.81904697,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.8552582263946533
    },
    {
      "auxiliary_loss_clip": 0.01168705,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.05199862,
      "balance_loss_mlp": 1.02105641,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 4.094721078842561,
      "language_loss": 0.72190464,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74387836,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 3.64172101020813
    },
    {
      "auxiliary_loss_clip": 0.01158946,
      "auxiliary_loss_mlp": 0.01026796,
      "balance_loss_clip": 1.04932523,
      "balance_loss_mlp": 1.01888919,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 1.9474398530699009,
      "language_loss": 0.68154836,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70340574,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.7549426555633545
    },
    {
      "auxiliary_loss_clip": 0.01158888,
      "auxiliary_loss_mlp": 0.00902095,
      "balance_loss_clip": 1.04929304,
      "balance_loss_mlp": 1.00088358,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.7643111075261122,
      "language_loss": 0.71906447,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.73967433,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.7027392387390137
    },
    {
      "auxiliary_loss_clip": 0.01071536,
      "auxiliary_loss_mlp": 0.01002075,
      "balance_loss_clip": 1.0205586,
      "balance_loss_mlp": 1.00095487,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.8984531093973447,
      "language_loss": 0.62696171,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64769781,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.1107146739959717
    },
    {
      "auxiliary_loss_clip": 0.01088451,
      "auxiliary_loss_mlp": 0.0100252,
      "balance_loss_clip": 1.02308762,
      "balance_loss_mlp": 1.00138128,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8625750444246258,
      "language_loss": 0.58424509,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60515475,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.894881248474121
    },
    {
      "auxiliary_loss_clip": 0.01175907,
      "auxiliary_loss_mlp": 0.0102599,
      "balance_loss_clip": 1.05207181,
      "balance_loss_mlp": 1.01758552,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 2.487503294312735,
      "language_loss": 0.74235505,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76437396,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.770016670227051
    },
    {
      "auxiliary_loss_clip": 0.0118183,
      "auxiliary_loss_mlp": 0.00901648,
      "balance_loss_clip": 1.05435312,
      "balance_loss_mlp": 1.0008769,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.7249134125263328,
      "language_loss": 0.66633999,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.6871748,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.78318190574646
    },
    {
      "auxiliary_loss_clip": 0.01177297,
      "auxiliary_loss_mlp": 0.01034461,
      "balance_loss_clip": 1.05451918,
      "balance_loss_mlp": 1.02603924,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 2.5973022287893697,
      "language_loss": 0.7734412,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79555875,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.6406304836273193
    },
    {
      "auxiliary_loss_clip": 0.01161043,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04927218,
      "balance_loss_mlp": 1.01805902,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 2.081360596263544,
      "language_loss": 0.77832538,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.80019444,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.7058701515197754
    },
    {
      "auxiliary_loss_clip": 0.0116334,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.05202448,
      "balance_loss_mlp": 1.01676893,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.2042972786294914,
      "language_loss": 0.80156374,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.82344562,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.7934436798095703
    },
    {
      "auxiliary_loss_clip": 0.01176059,
      "auxiliary_loss_mlp": 0.01025855,
      "balance_loss_clip": 1.05169129,
      "balance_loss_mlp": 1.01756406,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 2.070736889533384,
      "language_loss": 0.7202518,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74227095,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 4.190270900726318
    },
    {
      "auxiliary_loss_clip": 0.01173976,
      "auxiliary_loss_mlp": 0.01034513,
      "balance_loss_clip": 1.05068409,
      "balance_loss_mlp": 1.02619195,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.5725504600942901,
      "language_loss": 0.69584805,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.71793294,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.702831983566284
    },
    {
      "auxiliary_loss_clip": 0.01180026,
      "auxiliary_loss_mlp": 0.01028593,
      "balance_loss_clip": 1.05507755,
      "balance_loss_mlp": 1.02120209,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.1658316314088473,
      "language_loss": 0.78599799,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.80808413,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.563495635986328
    },
    {
      "auxiliary_loss_clip": 0.01155441,
      "auxiliary_loss_mlp": 0.01024979,
      "balance_loss_clip": 1.04678166,
      "balance_loss_mlp": 1.01732016,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 3.6862208843859867,
      "language_loss": 0.75178635,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77359051,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.6961588859558105
    },
    {
      "auxiliary_loss_clip": 0.01177697,
      "auxiliary_loss_mlp": 0.01023422,
      "balance_loss_clip": 1.05460024,
      "balance_loss_mlp": 1.01515436,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 2.100897509532622,
      "language_loss": 0.74400502,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76601619,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.828727960586548
    },
    {
      "auxiliary_loss_clip": 0.0115922,
      "auxiliary_loss_mlp": 0.01029121,
      "balance_loss_clip": 1.04854035,
      "balance_loss_mlp": 1.02122307,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 1.9450478079945683,
      "language_loss": 0.71735275,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73923618,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.6735944747924805
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.01028275,
      "balance_loss_clip": 1.05449867,
      "balance_loss_mlp": 1.02047837,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.5395668863543719,
      "language_loss": 0.77849066,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80057913,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 3.5706961154937744
    },
    {
      "auxiliary_loss_clip": 0.01159991,
      "auxiliary_loss_mlp": 0.01031174,
      "balance_loss_clip": 1.05292523,
      "balance_loss_mlp": 1.02330661,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 1.8710647876928783,
      "language_loss": 0.76223576,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78414738,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.636986255645752
    },
    {
      "auxiliary_loss_clip": 0.01179148,
      "auxiliary_loss_mlp": 0.01023586,
      "balance_loss_clip": 1.05277705,
      "balance_loss_mlp": 1.01595664,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.6202401087499225,
      "language_loss": 0.7426312,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76465857,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.6437323093414307
    },
    {
      "auxiliary_loss_clip": 0.01170795,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05112255,
      "balance_loss_mlp": 1.01892686,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.1267443056879394,
      "language_loss": 0.87257862,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89455748,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.621617078781128
    },
    {
      "auxiliary_loss_clip": 0.01165228,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.05444098,
      "balance_loss_mlp": 1.01801598,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.8910218051431513,
      "language_loss": 0.73726201,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.75917405,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 3.5777275562286377
    },
    {
      "auxiliary_loss_clip": 0.01164769,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.05065751,
      "balance_loss_mlp": 1.02158284,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.9629887906140235,
      "language_loss": 0.8088336,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.83078188,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.6787261962890625
    },
    {
      "auxiliary_loss_clip": 0.01159023,
      "auxiliary_loss_mlp": 0.01022447,
      "balance_loss_clip": 1.04789901,
      "balance_loss_mlp": 1.01472831,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.8214112717721604,
      "language_loss": 0.71867263,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74048734,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.6501305103302
    },
    {
      "auxiliary_loss_clip": 0.01161996,
      "auxiliary_loss_mlp": 0.01034117,
      "balance_loss_clip": 1.05651867,
      "balance_loss_mlp": 1.02603507,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.6883662601149156,
      "language_loss": 0.70852351,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.7304846,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.8223557472229004
    },
    {
      "auxiliary_loss_clip": 0.01168275,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.05055523,
      "balance_loss_mlp": 1.02225173,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.6439055721524112,
      "language_loss": 0.7760548,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.79804385,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.597261667251587
    },
    {
      "auxiliary_loss_clip": 0.0116631,
      "auxiliary_loss_mlp": 0.01031203,
      "balance_loss_clip": 1.05044127,
      "balance_loss_mlp": 1.02301967,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 2.4793075307956096,
      "language_loss": 0.78996122,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.81193638,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.640665054321289
    },
    {
      "auxiliary_loss_clip": 0.01178822,
      "auxiliary_loss_mlp": 0.01027479,
      "balance_loss_clip": 1.05176067,
      "balance_loss_mlp": 1.01978707,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 1.9710856054772465,
      "language_loss": 0.79938638,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.82144946,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.5417380332946777
    },
    {
      "auxiliary_loss_clip": 0.01169634,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.05264664,
      "balance_loss_mlp": 1.01718283,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.6871222021101047,
      "language_loss": 0.82039988,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.8423509,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 3.628077268600464
    },
    {
      "auxiliary_loss_clip": 0.01165689,
      "auxiliary_loss_mlp": 0.01025285,
      "balance_loss_clip": 1.05358946,
      "balance_loss_mlp": 1.01761961,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.143594673684743,
      "language_loss": 0.82910424,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85101402,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.699126720428467
    },
    {
      "auxiliary_loss_clip": 0.0116491,
      "auxiliary_loss_mlp": 0.01037251,
      "balance_loss_clip": 1.04901361,
      "balance_loss_mlp": 1.02919233,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 2.4214634793387027,
      "language_loss": 0.63297272,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65499437,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.7125120162963867
    },
    {
      "auxiliary_loss_clip": 0.01182199,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.0559659,
      "balance_loss_mlp": 1.02249897,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.5103012783888736,
      "language_loss": 0.71536088,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73748761,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.633359909057617
    },
    {
      "auxiliary_loss_clip": 0.01170786,
      "auxiliary_loss_mlp": 0.0090123,
      "balance_loss_clip": 1.05529261,
      "balance_loss_mlp": 1.0006783,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 2.007690427290749,
      "language_loss": 0.83417809,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85489833,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.6438610553741455
    },
    {
      "auxiliary_loss_clip": 0.01177155,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.05450237,
      "balance_loss_mlp": 1.0159874,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 3.1702831598737817,
      "language_loss": 0.86421013,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88621867,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.6610894203186035
    },
    {
      "auxiliary_loss_clip": 0.01163785,
      "auxiliary_loss_mlp": 0.01023865,
      "balance_loss_clip": 1.0503211,
      "balance_loss_mlp": 1.01645672,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 8.56722128031739,
      "language_loss": 0.82726562,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84914213,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.6464240550994873
    },
    {
      "auxiliary_loss_clip": 0.01166033,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.05282855,
      "balance_loss_mlp": 1.01882315,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.090656528018832,
      "language_loss": 0.79332536,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81524998,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.6871838569641113
    },
    {
      "auxiliary_loss_clip": 0.01162517,
      "auxiliary_loss_mlp": 0.01032371,
      "balance_loss_clip": 1.05033505,
      "balance_loss_mlp": 1.02337027,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 2.698647508111221,
      "language_loss": 0.70714927,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72909814,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.7511260509490967
    },
    {
      "auxiliary_loss_clip": 0.0116507,
      "auxiliary_loss_mlp": 0.01030201,
      "balance_loss_clip": 1.04714572,
      "balance_loss_mlp": 1.02271438,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.485911382683255,
      "language_loss": 0.65750557,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.67945832,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 3.758239984512329
    },
    {
      "auxiliary_loss_clip": 0.01160315,
      "auxiliary_loss_mlp": 0.0102811,
      "balance_loss_clip": 1.05237317,
      "balance_loss_mlp": 1.02020657,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 2.910410217184516,
      "language_loss": 0.78665769,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80854189,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.711430072784424
    },
    {
      "auxiliary_loss_clip": 0.01155697,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.05239403,
      "balance_loss_mlp": 1.01971602,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.747759101006784,
      "language_loss": 0.83402705,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85585928,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 2.6549432277679443
    },
    {
      "auxiliary_loss_clip": 0.01171636,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.05144846,
      "balance_loss_mlp": 1.02223396,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.40659200889047,
      "language_loss": 0.76256216,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78458136,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.653404712677002
    },
    {
      "auxiliary_loss_clip": 0.01159252,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.05028844,
      "balance_loss_mlp": 1.01913083,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.7773534071674204,
      "language_loss": 0.76502776,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78688979,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.635788679122925
    },
    {
      "auxiliary_loss_clip": 0.01083609,
      "auxiliary_loss_mlp": 0.01002139,
      "balance_loss_clip": 1.02231479,
      "balance_loss_mlp": 1.00095308,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7806680358824107,
      "language_loss": 0.57064176,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59149927,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.2167537212371826
    },
    {
      "auxiliary_loss_clip": 0.0118201,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.05487251,
      "balance_loss_mlp": 1.02141559,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.9462620591342537,
      "language_loss": 0.90575039,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92786342,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.657147169113159
    },
    {
      "auxiliary_loss_clip": 0.01168428,
      "auxiliary_loss_mlp": 0.01031796,
      "balance_loss_clip": 1.05005288,
      "balance_loss_mlp": 1.02349877,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.7274083078820934,
      "language_loss": 0.77862132,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.80062354,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 3.5859811305999756
    },
    {
      "auxiliary_loss_clip": 0.01163471,
      "auxiliary_loss_mlp": 0.01023371,
      "balance_loss_clip": 1.04869866,
      "balance_loss_mlp": 1.01586103,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.8038472928376508,
      "language_loss": 0.7598148,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78168321,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.6959309577941895
    },
    {
      "auxiliary_loss_clip": 0.01170631,
      "auxiliary_loss_mlp": 0.01021271,
      "balance_loss_clip": 1.05264783,
      "balance_loss_mlp": 1.01339698,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 2.7191141642280163,
      "language_loss": 0.7960484,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81796741,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.8422017097473145
    },
    {
      "auxiliary_loss_clip": 0.0117019,
      "auxiliary_loss_mlp": 0.01023272,
      "balance_loss_clip": 1.05106151,
      "balance_loss_mlp": 1.01537764,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 2.079275243247417,
      "language_loss": 0.69773692,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.71967149,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.6643199920654297
    },
    {
      "auxiliary_loss_clip": 0.01165397,
      "auxiliary_loss_mlp": 0.01024997,
      "balance_loss_clip": 1.05285835,
      "balance_loss_mlp": 1.01628304,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 2.608881013337188,
      "language_loss": 0.76381999,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78572392,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 3.6237192153930664
    },
    {
      "auxiliary_loss_clip": 0.01167996,
      "auxiliary_loss_mlp": 0.01028989,
      "balance_loss_clip": 1.05068302,
      "balance_loss_mlp": 1.02138925,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 2.0598028262961248,
      "language_loss": 0.86346591,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88543576,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.6267707347869873
    },
    {
      "auxiliary_loss_clip": 0.01186924,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.05680466,
      "balance_loss_mlp": 1.02397811,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 4.255165178607157,
      "language_loss": 0.72558236,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74777257,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.5909171104431152
    },
    {
      "auxiliary_loss_clip": 0.01158528,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.05032122,
      "balance_loss_mlp": 1.01798594,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 6.670320276902688,
      "language_loss": 0.75421143,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77605665,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.6556968688964844
    },
    {
      "auxiliary_loss_clip": 0.01185222,
      "auxiliary_loss_mlp": 0.01025256,
      "balance_loss_clip": 1.05572903,
      "balance_loss_mlp": 1.01713181,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.4494855574699503,
      "language_loss": 0.75302631,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77513111,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.58357572555542
    },
    {
      "auxiliary_loss_clip": 0.01164808,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.052001,
      "balance_loss_mlp": 1.02232575,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 1.91999838484093,
      "language_loss": 0.72043931,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.74238598,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.6917049884796143
    },
    {
      "auxiliary_loss_clip": 0.01161674,
      "auxiliary_loss_mlp": 0.01024891,
      "balance_loss_clip": 1.05310607,
      "balance_loss_mlp": 1.0169518,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.8702585144660677,
      "language_loss": 0.84061164,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.86247724,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 3.5838537216186523
    },
    {
      "auxiliary_loss_clip": 0.01179253,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.05289459,
      "balance_loss_mlp": 1.02067924,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.9571790950713013,
      "language_loss": 0.73812377,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.76020074,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.591728448867798
    },
    {
      "auxiliary_loss_clip": 0.01153013,
      "auxiliary_loss_mlp": 0.01032571,
      "balance_loss_clip": 1.04910862,
      "balance_loss_mlp": 1.02381504,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 5.602473733527524,
      "language_loss": 0.68103576,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70289159,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.842211961746216
    },
    {
      "auxiliary_loss_clip": 0.01077738,
      "auxiliary_loss_mlp": 0.0100173,
      "balance_loss_clip": 1.0202992,
      "balance_loss_mlp": 1.00051391,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.7765405936593192,
      "language_loss": 0.57445997,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59525466,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.075639486312866
    },
    {
      "auxiliary_loss_clip": 0.01162273,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 1.04809487,
      "balance_loss_mlp": 1.01779115,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.042311784749571,
      "language_loss": 0.75733435,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.77921504,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.675204277038574
    },
    {
      "auxiliary_loss_clip": 0.01151483,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.04507601,
      "balance_loss_mlp": 1.02406383,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 2.5796697702875764,
      "language_loss": 0.71335948,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73519158,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.739813804626465
    },
    {
      "auxiliary_loss_clip": 0.0117179,
      "auxiliary_loss_mlp": 0.01031853,
      "balance_loss_clip": 1.05299151,
      "balance_loss_mlp": 1.02370524,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.6498297490250367,
      "language_loss": 0.65130723,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67334366,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.7083842754364014
    },
    {
      "auxiliary_loss_clip": 0.01171107,
      "auxiliary_loss_mlp": 0.01025117,
      "balance_loss_clip": 1.0539484,
      "balance_loss_mlp": 1.01733851,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.6126290833711887,
      "language_loss": 0.76707214,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.78903443,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.6622233390808105
    },
    {
      "auxiliary_loss_clip": 0.01150997,
      "auxiliary_loss_mlp": 0.01026391,
      "balance_loss_clip": 1.04612815,
      "balance_loss_mlp": 1.01800466,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 2.204762668350973,
      "language_loss": 0.78339893,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.8051728,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.702353000640869
    },
    {
      "auxiliary_loss_clip": 0.01159634,
      "auxiliary_loss_mlp": 0.01027841,
      "balance_loss_clip": 1.04900455,
      "balance_loss_mlp": 1.0197165,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.5094105838388896,
      "language_loss": 0.79668498,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.81855977,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.668691873550415
    },
    {
      "auxiliary_loss_clip": 0.01072096,
      "auxiliary_loss_mlp": 0.01002888,
      "balance_loss_clip": 1.01838076,
      "balance_loss_mlp": 1.0015769,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.76650633591824,
      "language_loss": 0.58186102,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60261083,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 4.1379711627960205
    },
    {
      "auxiliary_loss_clip": 0.0117978,
      "auxiliary_loss_mlp": 0.01025818,
      "balance_loss_clip": 1.05344272,
      "balance_loss_mlp": 1.01789021,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.378838365892027,
      "language_loss": 0.67127609,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69333214,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 2.5758633613586426
    },
    {
      "auxiliary_loss_clip": 0.01165443,
      "auxiliary_loss_mlp": 0.01028932,
      "balance_loss_clip": 1.05053806,
      "balance_loss_mlp": 1.02022958,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 2.0385813870385316,
      "language_loss": 0.79926938,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.82121313,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.6825754642486572
    },
    {
      "auxiliary_loss_clip": 0.01151614,
      "auxiliary_loss_mlp": 0.01022122,
      "balance_loss_clip": 1.04870558,
      "balance_loss_mlp": 1.014153,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 3.0148518611738737,
      "language_loss": 0.75186205,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77359951,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.6495778560638428
    },
    {
      "auxiliary_loss_clip": 0.01158192,
      "auxiliary_loss_mlp": 0.01021384,
      "balance_loss_clip": 1.05305076,
      "balance_loss_mlp": 1.0136528,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.43881891947331,
      "language_loss": 0.82601857,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.84781432,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.7215614318847656
    },
    {
      "auxiliary_loss_clip": 0.01182067,
      "auxiliary_loss_mlp": 0.01022248,
      "balance_loss_clip": 1.05521154,
      "balance_loss_mlp": 1.01482713,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 1.7582426664949935,
      "language_loss": 0.79034007,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81238329,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.5413389205932617
    },
    {
      "auxiliary_loss_clip": 0.01167449,
      "auxiliary_loss_mlp": 0.01027526,
      "balance_loss_clip": 1.05249369,
      "balance_loss_mlp": 1.02030778,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.2869671938628158,
      "language_loss": 0.92049688,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94244659,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.6458353996276855
    },
    {
      "auxiliary_loss_clip": 0.01155614,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.04728317,
      "balance_loss_mlp": 1.01949692,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.7464283796587396,
      "language_loss": 0.82386327,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84569269,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 3.607095956802368
    },
    {
      "auxiliary_loss_clip": 0.01156,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.05093372,
      "balance_loss_mlp": 1.01965213,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 1.9037948758259697,
      "language_loss": 0.71069252,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.73253071,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.7290804386138916
    },
    {
      "auxiliary_loss_clip": 0.01169473,
      "auxiliary_loss_mlp": 0.01024162,
      "balance_loss_clip": 1.05007255,
      "balance_loss_mlp": 1.01607966,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.3820151270989096,
      "language_loss": 0.71439838,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.7363348,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.6488616466522217
    },
    {
      "auxiliary_loss_clip": 0.01165671,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.04914439,
      "balance_loss_mlp": 1.02027392,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.6535934112228048,
      "language_loss": 0.8190074,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.8409521,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.711078643798828
    },
    {
      "auxiliary_loss_clip": 0.01146439,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.0501039,
      "balance_loss_mlp": 1.02178812,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.6787000309822735,
      "language_loss": 0.84265161,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86441684,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 3.664579153060913
    },
    {
      "auxiliary_loss_clip": 0.01138874,
      "auxiliary_loss_mlp": 0.01030167,
      "balance_loss_clip": 1.04910159,
      "balance_loss_mlp": 1.02231073,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 3.426537394491224,
      "language_loss": 0.71543097,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.7371214,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.6557278633117676
    },
    {
      "auxiliary_loss_clip": 0.01157337,
      "auxiliary_loss_mlp": 0.01026321,
      "balance_loss_clip": 1.05275798,
      "balance_loss_mlp": 1.01797092,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.7056162292144839,
      "language_loss": 0.81142056,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83325714,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.705432891845703
    },
    {
      "auxiliary_loss_clip": 0.01172111,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 1.04891729,
      "balance_loss_mlp": 1.01997733,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 1.7941036967852635,
      "language_loss": 0.756697,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77869582,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.5857911109924316
    },
    {
      "auxiliary_loss_clip": 0.01153379,
      "auxiliary_loss_mlp": 0.01027099,
      "balance_loss_clip": 1.04860103,
      "balance_loss_mlp": 1.01949966,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 2.059049834298227,
      "language_loss": 0.80154097,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82334578,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.751237630844116
    },
    {
      "auxiliary_loss_clip": 0.01155926,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.04867768,
      "balance_loss_mlp": 1.01583242,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 2.022816424774525,
      "language_loss": 0.88856733,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.9103682,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.687370777130127
    },
    {
      "auxiliary_loss_clip": 0.01142442,
      "auxiliary_loss_mlp": 0.01025893,
      "balance_loss_clip": 1.04730463,
      "balance_loss_mlp": 1.01822197,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 2.0069118283486795,
      "language_loss": 0.83217454,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85385787,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 3.6278908252716064
    },
    {
      "auxiliary_loss_clip": 0.01181269,
      "auxiliary_loss_mlp": 0.0103078,
      "balance_loss_clip": 1.05340147,
      "balance_loss_mlp": 1.02258468,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 2.498684341733558,
      "language_loss": 0.79785061,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.8199712,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.685540199279785
    },
    {
      "auxiliary_loss_clip": 0.01153291,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04772592,
      "balance_loss_mlp": 1.01916265,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 2.043459949583313,
      "language_loss": 0.78437799,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.8061775,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.677860975265503
    },
    {
      "auxiliary_loss_clip": 0.01169574,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 1.04903126,
      "balance_loss_mlp": 1.0236783,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 2.5202510665814635,
      "language_loss": 0.77361828,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79563153,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.649582862854004
    },
    {
      "auxiliary_loss_clip": 0.01148525,
      "auxiliary_loss_mlp": 0.00901935,
      "balance_loss_clip": 1.047454,
      "balance_loss_mlp": 1.00087106,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 1.8665047275551476,
      "language_loss": 0.77752656,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.79803115,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.6796963214874268
    },
    {
      "auxiliary_loss_clip": 0.0115452,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.04760969,
      "balance_loss_mlp": 1.02087307,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.3470777165752623,
      "language_loss": 0.81025916,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83209383,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.73014497756958
    },
    {
      "auxiliary_loss_clip": 0.01085612,
      "auxiliary_loss_mlp": 0.01000772,
      "balance_loss_clip": 1.01975322,
      "balance_loss_mlp": 0.99954981,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7856645801444418,
      "language_loss": 0.60559189,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62645578,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.2992429733276367
    },
    {
      "auxiliary_loss_clip": 0.01159212,
      "auxiliary_loss_mlp": 0.01024465,
      "balance_loss_clip": 1.05447483,
      "balance_loss_mlp": 1.01678848,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.2931344053318488,
      "language_loss": 0.72856081,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.75039762,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.6804046630859375
    },
    {
      "auxiliary_loss_clip": 0.01141219,
      "auxiliary_loss_mlp": 0.00901547,
      "balance_loss_clip": 1.0436151,
      "balance_loss_mlp": 1.00077653,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 3.9984297474142276,
      "language_loss": 0.71432614,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.73475379,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.783292770385742
    },
    {
      "auxiliary_loss_clip": 0.01174556,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.05126762,
      "balance_loss_mlp": 1.0193913,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 2.072054269180914,
      "language_loss": 0.79920053,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82122314,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 3.520174741744995
    },
    {
      "auxiliary_loss_clip": 0.01149436,
      "auxiliary_loss_mlp": 0.01026647,
      "balance_loss_clip": 1.04768264,
      "balance_loss_mlp": 1.01873708,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 2.030547344328816,
      "language_loss": 0.7268461,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74860692,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.810866594314575
    },
    {
      "auxiliary_loss_clip": 0.01148875,
      "auxiliary_loss_mlp": 0.0103092,
      "balance_loss_clip": 1.05380273,
      "balance_loss_mlp": 1.02264118,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 5.016401581519858,
      "language_loss": 0.66648173,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68827963,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.7088465690612793
    },
    {
      "auxiliary_loss_clip": 0.01165866,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.05266476,
      "balance_loss_mlp": 1.02262068,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 3.3028271514119276,
      "language_loss": 0.69330531,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.7152679,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.65290904045105
    },
    {
      "auxiliary_loss_clip": 0.01135845,
      "auxiliary_loss_mlp": 0.01032267,
      "balance_loss_clip": 1.04557943,
      "balance_loss_mlp": 1.02348757,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.5441403124470399,
      "language_loss": 0.74413979,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76582086,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.835163116455078
    },
    {
      "auxiliary_loss_clip": 0.01170842,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 1.05235863,
      "balance_loss_mlp": 1.018332,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.1772170327183296,
      "language_loss": 0.79247189,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81444347,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.6861584186553955
    },
    {
      "auxiliary_loss_clip": 0.01176267,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.0510962,
      "balance_loss_mlp": 1.01904058,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.7907436200201206,
      "language_loss": 0.73380101,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75583208,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.635432243347168
    },
    {
      "auxiliary_loss_clip": 0.01114487,
      "auxiliary_loss_mlp": 0.01022442,
      "balance_loss_clip": 1.04640889,
      "balance_loss_mlp": 1.01546514,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 1.7473721300273772,
      "language_loss": 0.85558122,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.8769505,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 3.816056489944458
    },
    {
      "auxiliary_loss_clip": 0.01170848,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04960942,
      "balance_loss_mlp": 1.02269804,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 2.0575252526977397,
      "language_loss": 0.8819592,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90396619,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.8312573432922363
    },
    {
      "auxiliary_loss_clip": 0.01163708,
      "auxiliary_loss_mlp": 0.00901139,
      "balance_loss_clip": 1.04835188,
      "balance_loss_mlp": 1.00081134,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.7221884263961764,
      "language_loss": 0.74269068,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76333916,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.7123260498046875
    },
    {
      "auxiliary_loss_clip": 0.01156941,
      "auxiliary_loss_mlp": 0.01029916,
      "balance_loss_clip": 1.05259418,
      "balance_loss_mlp": 1.02178633,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.9254518785402848,
      "language_loss": 0.66268885,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68455744,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.6615586280822754
    },
    {
      "auxiliary_loss_clip": 0.01171054,
      "auxiliary_loss_mlp": 0.01024291,
      "balance_loss_clip": 1.05072427,
      "balance_loss_mlp": 1.01585746,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.794178522550808,
      "language_loss": 0.82658923,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.84854269,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 3.6110024452209473
    },
    {
      "auxiliary_loss_clip": 0.01165385,
      "auxiliary_loss_mlp": 0.01027693,
      "balance_loss_clip": 1.05138755,
      "balance_loss_mlp": 1.01959848,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 1.792624927803427,
      "language_loss": 0.79547989,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81741065,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.6361422538757324
    },
    {
      "auxiliary_loss_clip": 0.01166375,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.05123878,
      "balance_loss_mlp": 1.02080607,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.206566449164961,
      "language_loss": 0.85052824,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87248111,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.7568395137786865
    },
    {
      "auxiliary_loss_clip": 0.01171514,
      "auxiliary_loss_mlp": 0.00900545,
      "balance_loss_clip": 1.05271316,
      "balance_loss_mlp": 1.00078797,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.761137636147046,
      "language_loss": 0.87465662,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89537716,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.6404051780700684
    },
    {
      "auxiliary_loss_clip": 0.01184364,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.05668187,
      "balance_loss_mlp": 1.01677418,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 1.7494502877152107,
      "language_loss": 0.69554746,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71763796,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.611018180847168
    },
    {
      "auxiliary_loss_clip": 0.01172454,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.05481529,
      "balance_loss_mlp": 1.01960278,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.2133433989874605,
      "language_loss": 0.6174916,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.63949156,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.7215182781219482
    },
    {
      "auxiliary_loss_clip": 0.01171194,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.05262208,
      "balance_loss_mlp": 1.01990199,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.4247402715786635,
      "language_loss": 0.7050935,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.72708178,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.600543260574341
    },
    {
      "auxiliary_loss_clip": 0.01134554,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.04536033,
      "balance_loss_mlp": 1.01703513,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.416730789026542,
      "language_loss": 0.83334309,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85494065,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 3.74125337600708
    },
    {
      "auxiliary_loss_clip": 0.01131353,
      "auxiliary_loss_mlp": 0.00901518,
      "balance_loss_clip": 1.04616976,
      "balance_loss_mlp": 1.00069511,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 4.609385067232722,
      "language_loss": 0.76947707,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.78980583,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.9213931560516357
    },
    {
      "auxiliary_loss_clip": 0.01167454,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.0529387,
      "balance_loss_mlp": 1.01797009,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.646089731204185,
      "language_loss": 0.72262943,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74456394,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 3.081486701965332
    },
    {
      "auxiliary_loss_clip": 0.01170131,
      "auxiliary_loss_mlp": 0.01030205,
      "balance_loss_clip": 1.04979396,
      "balance_loss_mlp": 1.02231336,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 4.627022199669739,
      "language_loss": 0.89059222,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.91259551,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.573105812072754
    },
    {
      "auxiliary_loss_clip": 0.01127456,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.04701626,
      "balance_loss_mlp": 1.01649153,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.5048153359636072,
      "language_loss": 0.86070812,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.8822217,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.842103958129883
    },
    {
      "auxiliary_loss_clip": 0.01167838,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 1.0506587,
      "balance_loss_mlp": 1.02031851,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.356416903814694,
      "language_loss": 0.77183658,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79378963,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.59718918800354
    },
    {
      "auxiliary_loss_clip": 0.01179726,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.0552125,
      "balance_loss_mlp": 1.01841021,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 3.3259221619376955,
      "language_loss": 0.71922213,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.7412802,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.6260268688201904
    },
    {
      "auxiliary_loss_clip": 0.01172531,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.05049002,
      "balance_loss_mlp": 1.02133906,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.7075094664928923,
      "language_loss": 0.73373878,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75575656,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.61043119430542
    },
    {
      "auxiliary_loss_clip": 0.01080817,
      "auxiliary_loss_mlp": 0.0100156,
      "balance_loss_clip": 1.03323913,
      "balance_loss_mlp": 1.00050473,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8231414623622875,
      "language_loss": 0.60872966,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62955344,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 4.299105405807495
    },
    {
      "auxiliary_loss_clip": 0.01068714,
      "auxiliary_loss_mlp": 0.01007397,
      "balance_loss_clip": 1.03697538,
      "balance_loss_mlp": 1.00631809,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8435506284600703,
      "language_loss": 0.57816416,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59892517,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.1546168327331543
    },
    {
      "auxiliary_loss_clip": 0.01143622,
      "auxiliary_loss_mlp": 0.01028744,
      "balance_loss_clip": 1.04672122,
      "balance_loss_mlp": 1.02061367,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 3.7210248519146427,
      "language_loss": 0.75789845,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77962214,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.718416213989258
    },
    {
      "auxiliary_loss_clip": 0.01168845,
      "auxiliary_loss_mlp": 0.00901018,
      "balance_loss_clip": 1.05399561,
      "balance_loss_mlp": 1.00066054,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.6084613175166704,
      "language_loss": 0.70919037,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.72988892,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.591475009918213
    },
    {
      "auxiliary_loss_clip": 0.01165758,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.05201817,
      "balance_loss_mlp": 1.01940751,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 2.3465780183849554,
      "language_loss": 0.74506557,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76699519,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.6706185340881348
    },
    {
      "auxiliary_loss_clip": 0.0115866,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.04751897,
      "balance_loss_mlp": 1.01839399,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 1.9686401759619816,
      "language_loss": 0.78389645,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80574769,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.6694717407226562
    },
    {
      "auxiliary_loss_clip": 0.01159865,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 1.04984677,
      "balance_loss_mlp": 1.01856434,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.6401458150530512,
      "language_loss": 0.6962918,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.71815395,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.763461112976074
    },
    {
      "auxiliary_loss_clip": 0.01087112,
      "auxiliary_loss_mlp": 0.01002804,
      "balance_loss_clip": 1.02096009,
      "balance_loss_mlp": 1.00150418,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.7954714608103366,
      "language_loss": 0.51207167,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53297085,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 4.2049524784088135
    },
    {
      "auxiliary_loss_clip": 0.01143368,
      "auxiliary_loss_mlp": 0.01022803,
      "balance_loss_clip": 1.04665828,
      "balance_loss_mlp": 1.01444006,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.3770346962920716,
      "language_loss": 0.70153904,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72320074,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.6642873287200928
    },
    {
      "auxiliary_loss_clip": 0.01142957,
      "auxiliary_loss_mlp": 0.0102587,
      "balance_loss_clip": 1.04915023,
      "balance_loss_mlp": 1.01850867,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 2.2902010649302174,
      "language_loss": 0.71445423,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.73614252,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.7761244773864746
    },
    {
      "auxiliary_loss_clip": 0.01169825,
      "auxiliary_loss_mlp": 0.01027801,
      "balance_loss_clip": 1.05291271,
      "balance_loss_mlp": 1.01997519,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.5534796385166723,
      "language_loss": 0.78041017,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.80238646,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.7011473178863525
    },
    {
      "auxiliary_loss_clip": 0.01173528,
      "auxiliary_loss_mlp": 0.01025871,
      "balance_loss_clip": 1.05560446,
      "balance_loss_mlp": 1.01829576,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.5870845848839696,
      "language_loss": 0.85247588,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87446988,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 3.5395007133483887
    },
    {
      "auxiliary_loss_clip": 0.0118236,
      "auxiliary_loss_mlp": 0.010329,
      "balance_loss_clip": 1.05385625,
      "balance_loss_mlp": 1.0239892,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.697877775499689,
      "language_loss": 0.7419461,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76409864,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.580540895462036
    },
    {
      "auxiliary_loss_clip": 0.01162882,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.04833353,
      "balance_loss_mlp": 1.02280068,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.634569920375164,
      "language_loss": 0.78612411,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80805737,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.6698763370513916
    },
    {
      "auxiliary_loss_clip": 0.01153015,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.04324651,
      "balance_loss_mlp": 1.01786423,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.6616303964454457,
      "language_loss": 0.84268808,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.8644774,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.7227602005004883
    },
    {
      "auxiliary_loss_clip": 0.01172928,
      "auxiliary_loss_mlp": 0.01031425,
      "balance_loss_clip": 1.05419052,
      "balance_loss_mlp": 1.02330697,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.6832666945894705,
      "language_loss": 0.67292738,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69497085,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.673924207687378
    },
    {
      "auxiliary_loss_clip": 0.01179323,
      "auxiliary_loss_mlp": 0.01022745,
      "balance_loss_clip": 1.05416775,
      "balance_loss_mlp": 1.01522934,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.7104919881571397,
      "language_loss": 0.59431875,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61633939,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.728985071182251
    },
    {
      "auxiliary_loss_clip": 0.01160601,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.05100322,
      "balance_loss_mlp": 1.0167253,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.9966060351644914,
      "language_loss": 0.90991336,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93175948,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.6786513328552246
    },
    {
      "auxiliary_loss_clip": 0.01173221,
      "auxiliary_loss_mlp": 0.00901676,
      "balance_loss_clip": 1.05745578,
      "balance_loss_mlp": 1.00088882,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.7100316434736018,
      "language_loss": 0.76732415,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.78807306,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 3.5538992881774902
    },
    {
      "auxiliary_loss_clip": 0.01101804,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.03582263,
      "balance_loss_mlp": 1.02218628,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 2.3430156648302694,
      "language_loss": 0.76515794,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.7864759,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.9292187690734863
    },
    {
      "auxiliary_loss_clip": 0.01176164,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.0543437,
      "balance_loss_mlp": 1.01705551,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 2.2351137750788044,
      "language_loss": 0.78436911,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80638099,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.8155016899108887
    },
    {
      "auxiliary_loss_clip": 0.01134769,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.03855836,
      "balance_loss_mlp": 1.0188334,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 2.649068185909174,
      "language_loss": 0.730896,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.75251549,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.6614065170288086
    },
    {
      "auxiliary_loss_clip": 0.011249,
      "auxiliary_loss_mlp": 0.00900733,
      "balance_loss_clip": 1.04690289,
      "balance_loss_mlp": 1.00089157,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.1554780371078315,
      "language_loss": 0.75001359,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77026987,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.785566568374634
    },
    {
      "auxiliary_loss_clip": 0.01150029,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.04806161,
      "balance_loss_mlp": 1.02128863,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.796808356194203,
      "language_loss": 0.66942716,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.6912173,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.838651180267334
    },
    {
      "auxiliary_loss_clip": 0.01166245,
      "auxiliary_loss_mlp": 0.01025838,
      "balance_loss_clip": 1.0496068,
      "balance_loss_mlp": 1.01755285,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.6778403520482241,
      "language_loss": 0.75447845,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77639931,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.6216564178466797
    },
    {
      "auxiliary_loss_clip": 0.01183032,
      "auxiliary_loss_mlp": 0.01032355,
      "balance_loss_clip": 1.05687523,
      "balance_loss_mlp": 1.02430832,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.6889566983563375,
      "language_loss": 0.75595009,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77810395,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.6386258602142334
    },
    {
      "auxiliary_loss_clip": 0.01134629,
      "auxiliary_loss_mlp": 0.00901706,
      "balance_loss_clip": 1.04702735,
      "balance_loss_mlp": 1.00088263,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 1.906075572052515,
      "language_loss": 0.72386515,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74422848,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 4.017950534820557
    },
    {
      "auxiliary_loss_clip": 0.01142944,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04683018,
      "balance_loss_mlp": 1.02027321,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 1.8406673216801794,
      "language_loss": 0.71306455,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73478973,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 2.8208956718444824
    },
    {
      "auxiliary_loss_clip": 0.01152713,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.04701328,
      "balance_loss_mlp": 1.02193356,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.3086226464799706,
      "language_loss": 0.76708138,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.78891098,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.7649917602539062
    },
    {
      "auxiliary_loss_clip": 0.01162923,
      "auxiliary_loss_mlp": 0.01023495,
      "balance_loss_clip": 1.04878676,
      "balance_loss_mlp": 1.01607728,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.9123186166437776,
      "language_loss": 0.74022096,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76208508,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.7691283226013184
    },
    {
      "auxiliary_loss_clip": 0.01156941,
      "auxiliary_loss_mlp": 0.01025579,
      "balance_loss_clip": 1.04962873,
      "balance_loss_mlp": 1.01795864,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 4.112035646942526,
      "language_loss": 0.81612694,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83795214,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.783829689025879
    },
    {
      "auxiliary_loss_clip": 0.01158063,
      "auxiliary_loss_mlp": 0.01034264,
      "balance_loss_clip": 1.04652953,
      "balance_loss_mlp": 1.02578783,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.2721450261576344,
      "language_loss": 0.73644543,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75836873,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.8173820972442627
    },
    {
      "auxiliary_loss_clip": 0.01179274,
      "auxiliary_loss_mlp": 0.01026624,
      "balance_loss_clip": 1.05273199,
      "balance_loss_mlp": 1.0185113,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.7319618693260626,
      "language_loss": 0.69732141,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.71938038,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.7602477073669434
    },
    {
      "auxiliary_loss_clip": 0.01169387,
      "auxiliary_loss_mlp": 0.01021939,
      "balance_loss_clip": 1.05310297,
      "balance_loss_mlp": 1.01443458,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.6622358067793344,
      "language_loss": 0.75507295,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77698624,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.6742851734161377
    },
    {
      "auxiliary_loss_clip": 0.01147182,
      "auxiliary_loss_mlp": 0.01023317,
      "balance_loss_clip": 1.04630828,
      "balance_loss_mlp": 1.0155921,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.7177032435489838,
      "language_loss": 0.7088809,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.73058581,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 3.7350375652313232
    },
    {
      "auxiliary_loss_clip": 0.01151706,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 1.04931521,
      "balance_loss_mlp": 1.02097535,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.0989688331187213,
      "language_loss": 0.6208303,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64262772,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.9336984157562256
    },
    {
      "auxiliary_loss_clip": 0.01161928,
      "auxiliary_loss_mlp": 0.0103341,
      "balance_loss_clip": 1.05124807,
      "balance_loss_mlp": 1.02544129,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.806764496580941,
      "language_loss": 0.77459371,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79654711,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.7377209663391113
    },
    {
      "auxiliary_loss_clip": 0.01164676,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.0516417,
      "balance_loss_mlp": 1.01911867,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.9073152773620494,
      "language_loss": 0.81576657,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83768094,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.7412467002868652
    },
    {
      "auxiliary_loss_clip": 0.01180449,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 1.05273759,
      "balance_loss_mlp": 1.01699066,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 2.046017220964383,
      "language_loss": 0.80207241,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82413113,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 3.559643507003784
    },
    {
      "auxiliary_loss_clip": 0.01167305,
      "auxiliary_loss_mlp": 0.0102839,
      "balance_loss_clip": 1.05028522,
      "balance_loss_mlp": 1.02071881,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.9254347930332416,
      "language_loss": 0.72913301,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.75108993,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.68878436088562
    },
    {
      "auxiliary_loss_clip": 0.01171243,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.05367601,
      "balance_loss_mlp": 1.01871157,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.168987075029195,
      "language_loss": 0.75817394,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.78015453,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.628929615020752
    },
    {
      "auxiliary_loss_clip": 0.01172208,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.05139375,
      "balance_loss_mlp": 1.02168536,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 3.37269605230483,
      "language_loss": 0.84332114,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86533546,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.587747573852539
    },
    {
      "auxiliary_loss_clip": 0.01161472,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.05266643,
      "balance_loss_mlp": 1.02585793,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 3.1095354810799294,
      "language_loss": 0.71329021,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73524046,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.6920666694641113
    },
    {
      "auxiliary_loss_clip": 0.01165046,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.05175352,
      "balance_loss_mlp": 1.02199006,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.514934064754451,
      "language_loss": 0.73531932,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75727141,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.635699987411499
    },
    {
      "auxiliary_loss_clip": 0.01175603,
      "auxiliary_loss_mlp": 0.0102461,
      "balance_loss_clip": 1.05312753,
      "balance_loss_mlp": 1.0169301,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 2.7918875784517865,
      "language_loss": 0.74338728,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76538944,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.676115036010742
    },
    {
      "auxiliary_loss_clip": 0.01178695,
      "auxiliary_loss_mlp": 0.01027863,
      "balance_loss_clip": 1.05198002,
      "balance_loss_mlp": 1.02000678,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 3.9097944697938005,
      "language_loss": 0.85895997,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.88102555,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 3.449589252471924
    },
    {
      "auxiliary_loss_clip": 0.0114804,
      "auxiliary_loss_mlp": 0.01026501,
      "balance_loss_clip": 1.04698777,
      "balance_loss_mlp": 1.01807904,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.895786459992628,
      "language_loss": 0.85660112,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.8783465,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.715909719467163
    },
    {
      "auxiliary_loss_clip": 0.01172369,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.0531888,
      "balance_loss_mlp": 1.02235365,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 1.9560813102375723,
      "language_loss": 0.78310537,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80513185,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.6697726249694824
    },
    {
      "auxiliary_loss_clip": 0.01174141,
      "auxiliary_loss_mlp": 0.01026015,
      "balance_loss_clip": 1.05445361,
      "balance_loss_mlp": 1.01793277,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.764908332687808,
      "language_loss": 0.71957564,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74157721,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.794142007827759
    },
    {
      "auxiliary_loss_clip": 0.0116488,
      "auxiliary_loss_mlp": 0.01027136,
      "balance_loss_clip": 1.05293119,
      "balance_loss_mlp": 1.01969695,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 2.247886103626143,
      "language_loss": 0.76088387,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78280401,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.6628780364990234
    },
    {
      "auxiliary_loss_clip": 0.0113372,
      "auxiliary_loss_mlp": 0.0090124,
      "balance_loss_clip": 1.04472244,
      "balance_loss_mlp": 1.00063574,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.8358269789583734,
      "language_loss": 0.84612882,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86647832,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.713139057159424
    },
    {
      "auxiliary_loss_clip": 0.01163061,
      "auxiliary_loss_mlp": 0.0102454,
      "balance_loss_clip": 1.05054021,
      "balance_loss_mlp": 1.01747644,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.5304883916307297,
      "language_loss": 0.761841,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78371698,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.628765106201172
    },
    {
      "auxiliary_loss_clip": 0.011811,
      "auxiliary_loss_mlp": 0.01032467,
      "balance_loss_clip": 1.05228961,
      "balance_loss_mlp": 1.02450418,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.9749005118309089,
      "language_loss": 0.78271925,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80485487,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 3.467193126678467
    },
    {
      "auxiliary_loss_clip": 0.01132685,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.0474956,
      "balance_loss_mlp": 1.01633346,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.74628768082509,
      "language_loss": 0.90612549,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92769682,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.6735987663269043
    },
    {
      "auxiliary_loss_clip": 0.01168072,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04990554,
      "balance_loss_mlp": 1.01670408,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.5055537989650105,
      "language_loss": 0.82153213,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84345508,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 2.6115899085998535
    },
    {
      "auxiliary_loss_clip": 0.01164562,
      "auxiliary_loss_mlp": 0.01025485,
      "balance_loss_clip": 1.04952836,
      "balance_loss_mlp": 1.01733112,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.3430200276879285,
      "language_loss": 0.71951836,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.74141878,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.7698445320129395
    },
    {
      "auxiliary_loss_clip": 0.01144692,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04579353,
      "balance_loss_mlp": 1.01668966,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.7530646824738771,
      "language_loss": 0.70980978,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.73150253,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.7330851554870605
    },
    {
      "auxiliary_loss_clip": 0.01163122,
      "auxiliary_loss_mlp": 0.01025336,
      "balance_loss_clip": 1.04925132,
      "balance_loss_mlp": 1.0177778,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 2.284544118724519,
      "language_loss": 0.78084838,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80273294,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.6894195079803467
    },
    {
      "auxiliary_loss_clip": 0.01163776,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.0566473,
      "balance_loss_mlp": 1.01616096,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 1.7241680412005553,
      "language_loss": 0.66318828,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68506467,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.670654296875
    },
    {
      "auxiliary_loss_clip": 0.01160059,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.05054843,
      "balance_loss_mlp": 1.01837087,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.2197610479987078,
      "language_loss": 0.74614966,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76801038,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.605525016784668
    },
    {
      "auxiliary_loss_clip": 0.01172124,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 1.05450284,
      "balance_loss_mlp": 1.02458239,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.319059085671359,
      "language_loss": 0.67771947,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.69976121,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 3.5683186054229736
    },
    {
      "auxiliary_loss_clip": 0.01165264,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.05109572,
      "balance_loss_mlp": 1.01951444,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.8776216750774775,
      "language_loss": 0.72884291,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.75075799,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.579820156097412
    },
    {
      "auxiliary_loss_clip": 0.01158074,
      "auxiliary_loss_mlp": 0.01025419,
      "balance_loss_clip": 1.04894578,
      "balance_loss_mlp": 1.01685333,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 2.0600488415064233,
      "language_loss": 0.806301,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82813597,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.698934555053711
    },
    {
      "auxiliary_loss_clip": 0.01178871,
      "auxiliary_loss_mlp": 0.01025915,
      "balance_loss_clip": 1.05525589,
      "balance_loss_mlp": 1.01773691,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 2.534067961233386,
      "language_loss": 0.77386749,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79591537,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.5886104106903076
    },
    {
      "auxiliary_loss_clip": 0.01158805,
      "auxiliary_loss_mlp": 0.010222,
      "balance_loss_clip": 1.05075109,
      "balance_loss_mlp": 1.014678,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 1.8567494378623055,
      "language_loss": 0.69633603,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71814609,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 3.5827362537384033
    },
    {
      "auxiliary_loss_clip": 0.01164242,
      "auxiliary_loss_mlp": 0.01031269,
      "balance_loss_clip": 1.05254054,
      "balance_loss_mlp": 1.02356172,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.2687353911933443,
      "language_loss": 0.73087025,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75282538,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.7736880779266357
    },
    {
      "auxiliary_loss_clip": 0.01182958,
      "auxiliary_loss_mlp": 0.01026042,
      "balance_loss_clip": 1.05534863,
      "balance_loss_mlp": 1.01771784,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 2.53944456830676,
      "language_loss": 0.76038146,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.78247142,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.6042962074279785
    },
    {
      "auxiliary_loss_clip": 0.01165909,
      "auxiliary_loss_mlp": 0.01023034,
      "balance_loss_clip": 1.05108929,
      "balance_loss_mlp": 1.01528573,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.574718772225363,
      "language_loss": 0.80452377,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82641321,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.6811306476593018
    },
    {
      "auxiliary_loss_clip": 0.01166072,
      "auxiliary_loss_mlp": 0.01029954,
      "balance_loss_clip": 1.05185413,
      "balance_loss_mlp": 1.02172852,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 2.100704698559769,
      "language_loss": 0.70520741,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72716767,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.6471571922302246
    },
    {
      "auxiliary_loss_clip": 0.01129954,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.04204738,
      "balance_loss_mlp": 1.02329183,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 2.1722974087376032,
      "language_loss": 0.78077489,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80238485,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.7742085456848145
    },
    {
      "auxiliary_loss_clip": 0.01165707,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.05281806,
      "balance_loss_mlp": 1.01991749,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.925700873465855,
      "language_loss": 0.85427332,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87620795,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.6138949394226074
    },
    {
      "auxiliary_loss_clip": 0.01177831,
      "auxiliary_loss_mlp": 0.01030406,
      "balance_loss_clip": 1.0566467,
      "balance_loss_mlp": 1.02226996,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.9943357022529977,
      "language_loss": 0.74979162,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.77187401,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 3.576833724975586
    },
    {
      "auxiliary_loss_clip": 0.01175401,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.0540179,
      "balance_loss_mlp": 1.01889336,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 2.2115864379374277,
      "language_loss": 0.84427369,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86629546,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.6256256103515625
    },
    {
      "auxiliary_loss_clip": 0.01147833,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.04971671,
      "balance_loss_mlp": 1.02530575,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 1.9959495721454543,
      "language_loss": 0.68154359,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.703354,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.7768492698669434
    },
    {
      "auxiliary_loss_clip": 0.01168876,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 1.05138171,
      "balance_loss_mlp": 1.0206604,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 4.006938973060762,
      "language_loss": 0.71312165,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73510885,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.6994731426239014
    },
    {
      "auxiliary_loss_clip": 0.01180713,
      "auxiliary_loss_mlp": 0.01024187,
      "balance_loss_clip": 1.05489469,
      "balance_loss_mlp": 1.01688504,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 1.7393207708168332,
      "language_loss": 0.6081329,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.63018191,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.676311731338501
    },
    {
      "auxiliary_loss_clip": 0.01157506,
      "auxiliary_loss_mlp": 0.01028288,
      "balance_loss_clip": 1.04826367,
      "balance_loss_mlp": 1.02078986,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 2.1256465095435293,
      "language_loss": 0.82896501,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.85082293,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.6191024780273438
    },
    {
      "auxiliary_loss_clip": 0.01183447,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.05559564,
      "balance_loss_mlp": 1.01848185,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.8581479577819582,
      "language_loss": 0.79441631,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81651723,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.7370009422302246
    },
    {
      "auxiliary_loss_clip": 0.01163725,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.05068731,
      "balance_loss_mlp": 1.01739979,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.280246119792947,
      "language_loss": 0.75246894,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.7743566,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.6637399196624756
    },
    {
      "auxiliary_loss_clip": 0.01167694,
      "auxiliary_loss_mlp": 0.01024462,
      "balance_loss_clip": 1.05175138,
      "balance_loss_mlp": 1.01620102,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.3091086386825856,
      "language_loss": 0.71673745,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73865902,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 3.99440598487854
    },
    {
      "auxiliary_loss_clip": 0.01180458,
      "auxiliary_loss_mlp": 0.01028671,
      "balance_loss_clip": 1.05256057,
      "balance_loss_mlp": 1.02064228,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.2701677092877515,
      "language_loss": 0.84436834,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86645961,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.5603058338165283
    },
    {
      "auxiliary_loss_clip": 0.01169447,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.05340457,
      "balance_loss_mlp": 1.01734495,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 3.2651685372865087,
      "language_loss": 0.8729111,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89485717,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.5955276489257812
    },
    {
      "auxiliary_loss_clip": 0.01163028,
      "auxiliary_loss_mlp": 0.01030292,
      "balance_loss_clip": 1.05147338,
      "balance_loss_mlp": 1.02174497,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 2.2764204091700444,
      "language_loss": 0.66679764,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68873084,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.665245771408081
    },
    {
      "auxiliary_loss_clip": 0.01169035,
      "auxiliary_loss_mlp": 0.0102584,
      "balance_loss_clip": 1.0511148,
      "balance_loss_mlp": 1.01850843,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 1.9522562480220909,
      "language_loss": 0.82162231,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84357107,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.615105628967285
    },
    {
      "auxiliary_loss_clip": 0.01161418,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 1.05616093,
      "balance_loss_mlp": 1.02221274,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 18.534575868790903,
      "language_loss": 0.75421131,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77612543,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.6190357208251953
    },
    {
      "auxiliary_loss_clip": 0.01160709,
      "auxiliary_loss_mlp": 0.01020248,
      "balance_loss_clip": 1.05165577,
      "balance_loss_mlp": 1.01287496,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 5.7207397202529835,
      "language_loss": 0.71240675,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73421633,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.7261269092559814
    },
    {
      "auxiliary_loss_clip": 0.01132457,
      "auxiliary_loss_mlp": 0.01023953,
      "balance_loss_clip": 1.04476559,
      "balance_loss_mlp": 1.01634765,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 1.8902488497861225,
      "language_loss": 0.80216479,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82372898,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.7239348888397217
    },
    {
      "auxiliary_loss_clip": 0.01145507,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04511952,
      "balance_loss_mlp": 1.01812923,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 3.5176089741876453,
      "language_loss": 0.78012049,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.8018328,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 3.630777359008789
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01025105,
      "balance_loss_clip": 1.05390251,
      "balance_loss_mlp": 1.01785159,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.226069301646426,
      "language_loss": 0.67268586,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69460952,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.7131025791168213
    },
    {
      "auxiliary_loss_clip": 0.01178651,
      "auxiliary_loss_mlp": 0.01022029,
      "balance_loss_clip": 1.05361843,
      "balance_loss_mlp": 1.01422071,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.9470120910455104,
      "language_loss": 0.72410917,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74611592,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.597858190536499
    },
    {
      "auxiliary_loss_clip": 0.01157927,
      "auxiliary_loss_mlp": 0.010248,
      "balance_loss_clip": 1.05035377,
      "balance_loss_mlp": 1.01685429,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.1853866649695957,
      "language_loss": 0.64591491,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66774225,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.728111505508423
    },
    {
      "auxiliary_loss_clip": 0.01166179,
      "auxiliary_loss_mlp": 0.01034854,
      "balance_loss_clip": 1.04797471,
      "balance_loss_mlp": 1.02699828,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.8567036851731598,
      "language_loss": 0.8076725,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.82968283,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 3.6581039428710938
    },
    {
      "auxiliary_loss_clip": 0.0116048,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.04914927,
      "balance_loss_mlp": 1.02176094,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 1.8816301271202147,
      "language_loss": 0.76057279,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78248149,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.6980175971984863
    },
    {
      "auxiliary_loss_clip": 0.01161881,
      "auxiliary_loss_mlp": 0.01026078,
      "balance_loss_clip": 1.05179882,
      "balance_loss_mlp": 1.01838875,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 2.1513548723831373,
      "language_loss": 0.80954218,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83142179,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.684865951538086
    },
    {
      "auxiliary_loss_clip": 0.01167705,
      "auxiliary_loss_mlp": 0.01031844,
      "balance_loss_clip": 1.0535686,
      "balance_loss_mlp": 1.02376723,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.8976755380347332,
      "language_loss": 0.74041426,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76240981,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.691646099090576
    },
    {
      "auxiliary_loss_clip": 0.01116425,
      "auxiliary_loss_mlp": 0.01024282,
      "balance_loss_clip": 1.04276562,
      "balance_loss_mlp": 1.0171473,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 1.9016279499198625,
      "language_loss": 0.83859777,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86000484,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.7347795963287354
    },
    {
      "auxiliary_loss_clip": 0.01163372,
      "auxiliary_loss_mlp": 0.01030267,
      "balance_loss_clip": 1.05039918,
      "balance_loss_mlp": 1.02247667,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 16.77397976057563,
      "language_loss": 0.70706928,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72900569,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.6761868000030518
    },
    {
      "auxiliary_loss_clip": 0.01183186,
      "auxiliary_loss_mlp": 0.0103163,
      "balance_loss_clip": 1.05538845,
      "balance_loss_mlp": 1.02336836,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.6606738820194056,
      "language_loss": 0.7442317,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76637983,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.6290109157562256
    },
    {
      "auxiliary_loss_clip": 0.01171965,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.05663228,
      "balance_loss_mlp": 1.02134502,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 1.6962016061729024,
      "language_loss": 0.73354107,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75555736,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 3.6010942459106445
    },
    {
      "auxiliary_loss_clip": 0.01072882,
      "auxiliary_loss_mlp": 0.01008515,
      "balance_loss_clip": 1.02056813,
      "balance_loss_mlp": 1.00743008,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6859429604273498,
      "language_loss": 0.57212031,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59293425,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.353055715560913
    },
    {
      "auxiliary_loss_clip": 0.01163833,
      "auxiliary_loss_mlp": 0.01024278,
      "balance_loss_clip": 1.04999232,
      "balance_loss_mlp": 1.01707435,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 1.9052055307549747,
      "language_loss": 0.58537364,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.60725474,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.7240347862243652
    },
    {
      "auxiliary_loss_clip": 0.01179855,
      "auxiliary_loss_mlp": 0.01027045,
      "balance_loss_clip": 1.05405807,
      "balance_loss_mlp": 1.01980007,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 3.0811812184367966,
      "language_loss": 0.80165672,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.8237257,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.5854482650756836
    },
    {
      "auxiliary_loss_clip": 0.0116716,
      "auxiliary_loss_mlp": 0.01024985,
      "balance_loss_clip": 1.05174088,
      "balance_loss_mlp": 1.01769853,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 1.887096685382562,
      "language_loss": 0.80840731,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.83032876,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.6127281188964844
    },
    {
      "auxiliary_loss_clip": 0.01164581,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.05337501,
      "balance_loss_mlp": 1.02296615,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.770492497420236,
      "language_loss": 0.74922138,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77118272,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.655637502670288
    },
    {
      "auxiliary_loss_clip": 0.01162919,
      "auxiliary_loss_mlp": 0.01022086,
      "balance_loss_clip": 1.04985166,
      "balance_loss_mlp": 1.0145793,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 2.0234102349859118,
      "language_loss": 0.81804901,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.83989906,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.642033100128174
    },
    {
      "auxiliary_loss_clip": 0.01087035,
      "auxiliary_loss_mlp": 0.01003178,
      "balance_loss_clip": 1.02151775,
      "balance_loss_mlp": 1.00212336,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.8556920428074115,
      "language_loss": 0.60415179,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62505388,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.0850887298583984
    },
    {
      "auxiliary_loss_clip": 0.01174968,
      "auxiliary_loss_mlp": 0.01022658,
      "balance_loss_clip": 1.05004883,
      "balance_loss_mlp": 1.01422453,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 2.1948184098677173,
      "language_loss": 0.77393806,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79591429,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 3.5704922676086426
    },
    {
      "auxiliary_loss_clip": 0.0116351,
      "auxiliary_loss_mlp": 0.01029471,
      "balance_loss_clip": 1.04710913,
      "balance_loss_mlp": 1.02182698,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.93567826303063,
      "language_loss": 0.75493962,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77686942,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.738640785217285
    },
    {
      "auxiliary_loss_clip": 0.01149596,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.0477922,
      "balance_loss_mlp": 1.01726389,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 3.177959774409981,
      "language_loss": 0.74893701,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.77068216,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.713531732559204
    },
    {
      "auxiliary_loss_clip": 0.01170824,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.05195451,
      "balance_loss_mlp": 1.01746023,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 3.480905572403803,
      "language_loss": 0.82065189,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84261483,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.6225600242614746
    },
    {
      "auxiliary_loss_clip": 0.01159654,
      "auxiliary_loss_mlp": 0.01025589,
      "balance_loss_clip": 1.05202997,
      "balance_loss_mlp": 1.01801896,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 4.778969628652086,
      "language_loss": 0.69592988,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71778232,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.679560661315918
    },
    {
      "auxiliary_loss_clip": 0.01164629,
      "auxiliary_loss_mlp": 0.00901415,
      "balance_loss_clip": 1.05487621,
      "balance_loss_mlp": 1.00061214,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.248505749513447,
      "language_loss": 0.78797436,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.80863482,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.6243784427642822
    },
    {
      "auxiliary_loss_clip": 0.01146314,
      "auxiliary_loss_mlp": 0.01023687,
      "balance_loss_clip": 1.05047202,
      "balance_loss_mlp": 1.01609898,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 2.0520395715402557,
      "language_loss": 0.74249482,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76419485,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.6718764305114746
    },
    {
      "auxiliary_loss_clip": 0.01170415,
      "auxiliary_loss_mlp": 0.01029465,
      "balance_loss_clip": 1.05298877,
      "balance_loss_mlp": 1.02261305,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.0894283574957413,
      "language_loss": 0.84733933,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.86933815,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.6111643314361572
    },
    {
      "auxiliary_loss_clip": 0.01073571,
      "auxiliary_loss_mlp": 0.01003155,
      "balance_loss_clip": 1.01897788,
      "balance_loss_mlp": 1.0019387,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7496464129540636,
      "language_loss": 0.5226568,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54342401,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 4.142684459686279
    },
    {
      "auxiliary_loss_clip": 0.01181298,
      "auxiliary_loss_mlp": 0.01023346,
      "balance_loss_clip": 1.05342412,
      "balance_loss_mlp": 1.01554978,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.110194905385653,
      "language_loss": 0.65550345,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.6775499,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.742899179458618
    },
    {
      "auxiliary_loss_clip": 0.01152232,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.04925966,
      "balance_loss_mlp": 1.02057695,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 2.906316988131187,
      "language_loss": 0.84897977,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.87078846,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.668243408203125
    },
    {
      "auxiliary_loss_clip": 0.01153939,
      "auxiliary_loss_mlp": 0.01023167,
      "balance_loss_clip": 1.04530692,
      "balance_loss_mlp": 1.01535249,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.680968398589915,
      "language_loss": 0.77916783,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.8009389,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.6737003326416016
    },
    {
      "auxiliary_loss_clip": 0.01173717,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.0501914,
      "balance_loss_mlp": 1.02073121,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.731932558535477,
      "language_loss": 0.76560277,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78762949,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 3.6623916625976562
    },
    {
      "auxiliary_loss_clip": 0.01159919,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.04626548,
      "balance_loss_mlp": 1.02060163,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.757621289180301,
      "language_loss": 0.71331918,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73519433,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.775111675262451
    },
    {
      "auxiliary_loss_clip": 0.01159504,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 1.05346382,
      "balance_loss_mlp": 1.01915908,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 1.869257162052125,
      "language_loss": 0.72346467,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74533218,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.7152183055877686
    },
    {
      "auxiliary_loss_clip": 0.01157597,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.04897141,
      "balance_loss_mlp": 1.02207172,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 2.0454779726142673,
      "language_loss": 0.73152041,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75339502,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.731381416320801
    },
    {
      "auxiliary_loss_clip": 0.01151503,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.04896092,
      "balance_loss_mlp": 1.02180851,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.6889230526647447,
      "language_loss": 0.80208749,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82389504,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.658989191055298
    },
    {
      "auxiliary_loss_clip": 0.01182687,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.0521431,
      "balance_loss_mlp": 1.01854587,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.606703286180611,
      "language_loss": 0.80638754,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.82848239,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.6012887954711914
    },
    {
      "auxiliary_loss_clip": 0.01137268,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.04906523,
      "balance_loss_mlp": 1.02099717,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 2.2024598213911872,
      "language_loss": 0.81970203,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.84136236,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.698957920074463
    },
    {
      "auxiliary_loss_clip": 0.01156077,
      "auxiliary_loss_mlp": 0.01023577,
      "balance_loss_clip": 1.04675901,
      "balance_loss_mlp": 1.01560807,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 1.8109623582546743,
      "language_loss": 0.9162451,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93804169,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 3.658328056335449
    },
    {
      "auxiliary_loss_clip": 0.01161791,
      "auxiliary_loss_mlp": 0.01026963,
      "balance_loss_clip": 1.05062175,
      "balance_loss_mlp": 1.01934004,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 2.014043311551311,
      "language_loss": 0.77445614,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79634368,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.7242915630340576
    },
    {
      "auxiliary_loss_clip": 0.01163431,
      "auxiliary_loss_mlp": 0.01026006,
      "balance_loss_clip": 1.05064845,
      "balance_loss_mlp": 1.01747072,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.268640711268469,
      "language_loss": 0.78735304,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.80924743,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.686289072036743
    },
    {
      "auxiliary_loss_clip": 0.01158678,
      "auxiliary_loss_mlp": 0.01024549,
      "balance_loss_clip": 1.05174804,
      "balance_loss_mlp": 1.01697874,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 2.028409655233221,
      "language_loss": 0.76180804,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78364027,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.683128833770752
    },
    {
      "auxiliary_loss_clip": 0.01176588,
      "auxiliary_loss_mlp": 0.01024588,
      "balance_loss_clip": 1.05385339,
      "balance_loss_mlp": 1.01732492,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.943643056419042,
      "language_loss": 0.79300296,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81501472,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.6281464099884033
    },
    {
      "auxiliary_loss_clip": 0.01085326,
      "auxiliary_loss_mlp": 0.01005104,
      "balance_loss_clip": 1.02122629,
      "balance_loss_mlp": 1.00410235,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7760584764401333,
      "language_loss": 0.54374063,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56464493,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.185116767883301
    },
    {
      "auxiliary_loss_clip": 0.01152843,
      "auxiliary_loss_mlp": 0.01030276,
      "balance_loss_clip": 1.04817772,
      "balance_loss_mlp": 1.02230132,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.437369572284128,
      "language_loss": 0.71953219,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74136341,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.6782500743865967
    },
    {
      "auxiliary_loss_clip": 0.01178893,
      "auxiliary_loss_mlp": 0.01023237,
      "balance_loss_clip": 1.05553401,
      "balance_loss_mlp": 1.01624227,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.6129692558688917,
      "language_loss": 0.71215165,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73417294,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 3.661655902862549
    },
    {
      "auxiliary_loss_clip": 0.01180194,
      "auxiliary_loss_mlp": 0.00900656,
      "balance_loss_clip": 1.05437136,
      "balance_loss_mlp": 1.00060058,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 2.150870042393098,
      "language_loss": 0.77030921,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.79111767,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.550741672515869
    },
    {
      "auxiliary_loss_clip": 0.01165458,
      "auxiliary_loss_mlp": 0.01026588,
      "balance_loss_clip": 1.0523653,
      "balance_loss_mlp": 1.01961434,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.9200590730493474,
      "language_loss": 0.80900693,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83092737,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.6010777950286865
    },
    {
      "auxiliary_loss_clip": 0.01152377,
      "auxiliary_loss_mlp": 0.01020971,
      "balance_loss_clip": 1.04418707,
      "balance_loss_mlp": 1.01334763,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 2.102812591414391,
      "language_loss": 0.60489035,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62662387,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.611213445663452
    },
    {
      "auxiliary_loss_clip": 0.0112017,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.04247332,
      "balance_loss_mlp": 1.01980329,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 2.3968638439410785,
      "language_loss": 0.76310027,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78457296,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.8202314376831055
    },
    {
      "auxiliary_loss_clip": 0.01163269,
      "auxiliary_loss_mlp": 0.01029822,
      "balance_loss_clip": 1.0488472,
      "balance_loss_mlp": 1.02203739,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.3831267776723517,
      "language_loss": 0.65266323,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67459416,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.654372215270996
    },
    {
      "auxiliary_loss_clip": 0.01155656,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.04709113,
      "balance_loss_mlp": 1.01715648,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.2349300322557935,
      "language_loss": 0.68880975,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.710621,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.6213219165802
    },
    {
      "auxiliary_loss_clip": 0.01159009,
      "auxiliary_loss_mlp": 0.01024555,
      "balance_loss_clip": 1.04742694,
      "balance_loss_mlp": 1.01656163,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 3.0462123437669177,
      "language_loss": 0.73296344,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75479913,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.59679913520813
    },
    {
      "auxiliary_loss_clip": 0.01141049,
      "auxiliary_loss_mlp": 0.0102746,
      "balance_loss_clip": 1.04487264,
      "balance_loss_mlp": 1.01959825,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 2.0261429600545986,
      "language_loss": 0.71389914,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.7355842,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 3.687408924102783
    },
    {
      "auxiliary_loss_clip": 0.0115183,
      "auxiliary_loss_mlp": 0.01024234,
      "balance_loss_clip": 1.04578054,
      "balance_loss_mlp": 1.01692617,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.485888099987769,
      "language_loss": 0.76671588,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78847659,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.6860320568084717
    },
    {
      "auxiliary_loss_clip": 0.0115353,
      "auxiliary_loss_mlp": 0.01025824,
      "balance_loss_clip": 1.04701042,
      "balance_loss_mlp": 1.01781857,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 2.066626697439828,
      "language_loss": 0.78584766,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80764121,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.6684486865997314
    },
    {
      "auxiliary_loss_clip": 0.01116412,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.0439322,
      "balance_loss_mlp": 1.02009511,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 2.189528036431926,
      "language_loss": 0.69846815,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71990424,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.893887758255005
    },
    {
      "auxiliary_loss_clip": 0.01172534,
      "auxiliary_loss_mlp": 0.01023859,
      "balance_loss_clip": 1.0515821,
      "balance_loss_mlp": 1.01573491,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 1.90484301973425,
      "language_loss": 0.65009862,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67206258,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 3.952897310256958
    },
    {
      "auxiliary_loss_clip": 0.01168679,
      "auxiliary_loss_mlp": 0.01020884,
      "balance_loss_clip": 1.04880178,
      "balance_loss_mlp": 1.01331449,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 1.8865953041812578,
      "language_loss": 0.78186202,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80375773,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.9976611137390137
    },
    {
      "auxiliary_loss_clip": 0.01163782,
      "auxiliary_loss_mlp": 0.01024912,
      "balance_loss_clip": 1.04965043,
      "balance_loss_mlp": 1.01713908,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 2.2350966257621128,
      "language_loss": 0.8194837,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.8413707,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.631955146789551
    },
    {
      "auxiliary_loss_clip": 0.01136255,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.04333079,
      "balance_loss_mlp": 1.01444745,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 1.881139333107867,
      "language_loss": 0.84789872,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.86948359,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.6736271381378174
    },
    {
      "auxiliary_loss_clip": 0.01141716,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.04723573,
      "balance_loss_mlp": 1.01930296,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 2.732438947714986,
      "language_loss": 0.89994466,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92162901,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.777024984359741
    },
    {
      "auxiliary_loss_clip": 0.01157308,
      "auxiliary_loss_mlp": 0.01025268,
      "balance_loss_clip": 1.04796791,
      "balance_loss_mlp": 1.0182023,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.993638105191719,
      "language_loss": 0.7653054,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78713119,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.619692325592041
    },
    {
      "auxiliary_loss_clip": 0.01148569,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.04987943,
      "balance_loss_mlp": 1.02115166,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.042040458498317,
      "language_loss": 0.75478876,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77656865,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 3.665898084640503
    },
    {
      "auxiliary_loss_clip": 0.01150878,
      "auxiliary_loss_mlp": 0.01035514,
      "balance_loss_clip": 1.04699183,
      "balance_loss_mlp": 1.02629936,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.014258524198026,
      "language_loss": 0.78280097,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80466491,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.693659782409668
    },
    {
      "auxiliary_loss_clip": 0.01113082,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.04072833,
      "balance_loss_mlp": 1.01743627,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 3.096192314205623,
      "language_loss": 0.73472142,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75610578,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 2.976165533065796
    },
    {
      "auxiliary_loss_clip": 0.01159707,
      "auxiliary_loss_mlp": 0.01024406,
      "balance_loss_clip": 1.05028176,
      "balance_loss_mlp": 1.01690233,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 3.33084491762574,
      "language_loss": 0.77190483,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79374593,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 3.166377305984497
    },
    {
      "auxiliary_loss_clip": 0.01167496,
      "auxiliary_loss_mlp": 0.01021732,
      "balance_loss_clip": 1.04934311,
      "balance_loss_mlp": 1.01373899,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.1152601811934653,
      "language_loss": 0.67306709,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69495928,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.6340315341949463
    },
    {
      "auxiliary_loss_clip": 0.01162675,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.0488174,
      "balance_loss_mlp": 1.01682913,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.8942094080836152,
      "language_loss": 0.84212697,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86398935,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.6730735301971436
    },
    {
      "auxiliary_loss_clip": 0.01167602,
      "auxiliary_loss_mlp": 0.00900943,
      "balance_loss_clip": 1.04824626,
      "balance_loss_mlp": 1.00081587,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.917277184862803,
      "language_loss": 0.72320867,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.7438941,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.6734726428985596
    },
    {
      "auxiliary_loss_clip": 0.01178262,
      "auxiliary_loss_mlp": 0.01030577,
      "balance_loss_clip": 1.05126452,
      "balance_loss_mlp": 1.02249432,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.472375958533196,
      "language_loss": 0.76524025,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78732866,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.554060935974121
    },
    {
      "auxiliary_loss_clip": 0.01164594,
      "auxiliary_loss_mlp": 0.01028247,
      "balance_loss_clip": 1.05082679,
      "balance_loss_mlp": 1.01970589,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.823821025291695,
      "language_loss": 0.68551064,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.70743901,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.670772075653076
    },
    {
      "auxiliary_loss_clip": 0.0113289,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 1.04094625,
      "balance_loss_mlp": 1.01821709,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 3.548055509653656,
      "language_loss": 0.81221545,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83380163,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 4.007980823516846
    },
    {
      "auxiliary_loss_clip": 0.01142131,
      "auxiliary_loss_mlp": 0.0103377,
      "balance_loss_clip": 1.04543161,
      "balance_loss_mlp": 1.02532411,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 3.20706997042579,
      "language_loss": 0.747711,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.76946998,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.714980125427246
    },
    {
      "auxiliary_loss_clip": 0.01159075,
      "auxiliary_loss_mlp": 0.01025779,
      "balance_loss_clip": 1.05068302,
      "balance_loss_mlp": 1.01795316,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 2.0745675817623837,
      "language_loss": 0.73921299,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76106155,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.6932342052459717
    },
    {
      "auxiliary_loss_clip": 0.0118503,
      "auxiliary_loss_mlp": 0.01033386,
      "balance_loss_clip": 1.05661464,
      "balance_loss_mlp": 1.02569759,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.4939309811690105,
      "language_loss": 0.76094794,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.7831322,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.572903871536255
    },
    {
      "auxiliary_loss_clip": 0.01154223,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.04715168,
      "balance_loss_mlp": 1.02085161,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.795156490149913,
      "language_loss": 0.79096162,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81278348,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.7099452018737793
    },
    {
      "auxiliary_loss_clip": 0.0115887,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.05144548,
      "balance_loss_mlp": 1.01872754,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 3.6610191068101936,
      "language_loss": 0.78082848,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80268157,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.745751142501831
    },
    {
      "auxiliary_loss_clip": 0.01169509,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.05370331,
      "balance_loss_mlp": 1.01875043,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 2.1536068111999978,
      "language_loss": 0.72101116,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.7429691,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.6910147666931152
    },
    {
      "auxiliary_loss_clip": 0.01033357,
      "auxiliary_loss_mlp": 0.01006144,
      "balance_loss_clip": 1.01569521,
      "balance_loss_mlp": 1.00514281,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7431934873210381,
      "language_loss": 0.52734965,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54774469,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.2301924228668213
    },
    {
      "auxiliary_loss_clip": 0.01158573,
      "auxiliary_loss_mlp": 0.01033246,
      "balance_loss_clip": 1.05349827,
      "balance_loss_mlp": 1.02547979,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 2.024322816004001,
      "language_loss": 0.79102463,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.8129428,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 3.8036696910858154
    },
    {
      "auxiliary_loss_clip": 0.01152338,
      "auxiliary_loss_mlp": 0.01030584,
      "balance_loss_clip": 1.05005145,
      "balance_loss_mlp": 1.02313602,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 1.8266097223211482,
      "language_loss": 0.73708081,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75890994,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.7916860580444336
    },
    {
      "auxiliary_loss_clip": 0.01179153,
      "auxiliary_loss_mlp": 0.01024934,
      "balance_loss_clip": 1.05202651,
      "balance_loss_mlp": 1.01688719,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 3.182502971990268,
      "language_loss": 0.62506956,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.6471104,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.6217870712280273
    },
    {
      "auxiliary_loss_clip": 0.01167483,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.05201948,
      "balance_loss_mlp": 1.0164659,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.891667905995606,
      "language_loss": 0.82983947,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85175866,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 3.59820556640625
    },
    {
      "auxiliary_loss_clip": 0.01173498,
      "auxiliary_loss_mlp": 0.01031493,
      "balance_loss_clip": 1.05440462,
      "balance_loss_mlp": 1.02398264,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.244948627543489,
      "language_loss": 0.80704677,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.82909667,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.582047462463379
    },
    {
      "auxiliary_loss_clip": 0.01152134,
      "auxiliary_loss_mlp": 0.01024661,
      "balance_loss_clip": 1.04405951,
      "balance_loss_mlp": 1.01665652,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 4.015792326284053,
      "language_loss": 0.7218678,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74363577,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.6937944889068604
    },
    {
      "auxiliary_loss_clip": 0.01158589,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.04889524,
      "balance_loss_mlp": 1.01632237,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 2.069129832531163,
      "language_loss": 0.87060755,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89243448,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.63566517829895
    },
    {
      "auxiliary_loss_clip": 0.01164665,
      "auxiliary_loss_mlp": 0.01024751,
      "balance_loss_clip": 1.05143213,
      "balance_loss_mlp": 1.017169,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.6178888486737055,
      "language_loss": 0.80273497,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82462907,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.67911434173584
    },
    {
      "auxiliary_loss_clip": 0.01177391,
      "auxiliary_loss_mlp": 0.01031923,
      "balance_loss_clip": 1.05312014,
      "balance_loss_mlp": 1.02365613,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 2.167747977819108,
      "language_loss": 0.71669465,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73878777,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.703991413116455
    },
    {
      "auxiliary_loss_clip": 0.01160861,
      "auxiliary_loss_mlp": 0.00901481,
      "balance_loss_clip": 1.05031979,
      "balance_loss_mlp": 1.00093961,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 1.8674276095835567,
      "language_loss": 0.72547662,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74610001,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.8134844303131104
    },
    {
      "auxiliary_loss_clip": 0.0107387,
      "auxiliary_loss_mlp": 0.01000865,
      "balance_loss_clip": 1.01872325,
      "balance_loss_mlp": 0.99995941,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8496560629199845,
      "language_loss": 0.63314927,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65389657,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 4.089256048202515
    },
    {
      "auxiliary_loss_clip": 0.01172834,
      "auxiliary_loss_mlp": 0.01025055,
      "balance_loss_clip": 1.05281973,
      "balance_loss_mlp": 1.01714587,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.8288289137211016,
      "language_loss": 0.74343169,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76541066,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.6180636882781982
    },
    {
      "auxiliary_loss_clip": 0.01171301,
      "auxiliary_loss_mlp": 0.01029721,
      "balance_loss_clip": 1.0514369,
      "balance_loss_mlp": 1.02218151,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.4451892591134088,
      "language_loss": 0.72808087,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.75009108,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.639573812484741
    },
    {
      "auxiliary_loss_clip": 0.01169003,
      "auxiliary_loss_mlp": 0.01024356,
      "balance_loss_clip": 1.05272269,
      "balance_loss_mlp": 1.01642275,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 3.057682616286091,
      "language_loss": 0.78247488,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.80440849,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.5738494396209717
    },
    {
      "auxiliary_loss_clip": 0.01168881,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.05079865,
      "balance_loss_mlp": 1.02247119,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.7518138546297326,
      "language_loss": 0.78788018,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80987144,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.6844661235809326
    },
    {
      "auxiliary_loss_clip": 0.0117206,
      "auxiliary_loss_mlp": 0.01022687,
      "balance_loss_clip": 1.05351484,
      "balance_loss_mlp": 1.01436567,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.6015639300498892,
      "language_loss": 0.63850796,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.66045547,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.6502890586853027
    },
    {
      "auxiliary_loss_clip": 0.01148923,
      "auxiliary_loss_mlp": 0.01024477,
      "balance_loss_clip": 1.04739714,
      "balance_loss_mlp": 1.016716,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 2.2238911461688784,
      "language_loss": 0.78853238,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81026638,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.6418068408966064
    },
    {
      "auxiliary_loss_clip": 0.01141431,
      "auxiliary_loss_mlp": 0.01023852,
      "balance_loss_clip": 1.04238486,
      "balance_loss_mlp": 1.01615667,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.979865185002554,
      "language_loss": 0.73637974,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75803262,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.7978060245513916
    },
    {
      "auxiliary_loss_clip": 0.01167794,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.05139518,
      "balance_loss_mlp": 1.02139592,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.7877312716388687,
      "language_loss": 0.71751058,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.73947364,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 3.488892078399658
    },
    {
      "auxiliary_loss_clip": 0.0112261,
      "auxiliary_loss_mlp": 0.01024336,
      "balance_loss_clip": 1.04052961,
      "balance_loss_mlp": 1.01677179,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.0452047986774198,
      "language_loss": 0.75266302,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77413249,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.804054021835327
    },
    {
      "auxiliary_loss_clip": 0.01180179,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.05369544,
      "balance_loss_mlp": 1.01793396,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.462672432537807,
      "language_loss": 0.80432558,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82639039,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.5384416580200195
    },
    {
      "auxiliary_loss_clip": 0.01141026,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04889882,
      "balance_loss_mlp": 1.01964486,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.9687620909640489,
      "language_loss": 0.7417087,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76339811,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.717902898788452
    },
    {
      "auxiliary_loss_clip": 0.01137332,
      "auxiliary_loss_mlp": 0.0102439,
      "balance_loss_clip": 1.04801822,
      "balance_loss_mlp": 1.01677227,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.9687846209852957,
      "language_loss": 0.78520596,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.80682319,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.952785015106201
    },
    {
      "auxiliary_loss_clip": 0.01126,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04447865,
      "balance_loss_mlp": 1.02062702,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.7776360529339492,
      "language_loss": 0.80179346,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82333475,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.7425947189331055
    },
    {
      "auxiliary_loss_clip": 0.01166818,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.05214286,
      "balance_loss_mlp": 1.02034998,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4634526371527292,
      "language_loss": 0.7566129,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.7785604,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.6323657035827637
    },
    {
      "auxiliary_loss_clip": 0.01170038,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.04975593,
      "balance_loss_mlp": 1.01652181,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.2200441681097534,
      "language_loss": 0.70552349,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72746789,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.6781928539276123
    },
    {
      "auxiliary_loss_clip": 0.01148572,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.04357791,
      "balance_loss_mlp": 1.02033782,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.154645013203959,
      "language_loss": 0.66167736,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68344784,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 3.490013599395752
    },
    {
      "auxiliary_loss_clip": 0.01166773,
      "auxiliary_loss_mlp": 0.01025408,
      "balance_loss_clip": 1.04993916,
      "balance_loss_mlp": 1.01778424,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 1.757894165712478,
      "language_loss": 0.81817198,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.84009373,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.504342555999756
    },
    {
      "auxiliary_loss_clip": 0.01151909,
      "auxiliary_loss_mlp": 0.01023097,
      "balance_loss_clip": 1.04942465,
      "balance_loss_mlp": 1.0157063,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.0550126999471012,
      "language_loss": 0.78057766,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.80232775,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.515509605407715
    },
    {
      "auxiliary_loss_clip": 0.01171977,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.05189824,
      "balance_loss_mlp": 1.01883852,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 1.9474195323737695,
      "language_loss": 0.73225898,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75424838,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 3.5731797218322754
    },
    {
      "auxiliary_loss_clip": 0.01145091,
      "auxiliary_loss_mlp": 0.01022678,
      "balance_loss_clip": 1.04438925,
      "balance_loss_mlp": 1.0144769,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.4370234378089073,
      "language_loss": 0.8408854,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86256313,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.8113722801208496
    },
    {
      "auxiliary_loss_clip": 0.011459,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.0465709,
      "balance_loss_mlp": 1.02002358,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 2.1021662277938997,
      "language_loss": 0.81315547,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83489048,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.755591630935669
    },
    {
      "auxiliary_loss_clip": 0.01171584,
      "auxiliary_loss_mlp": 0.01030962,
      "balance_loss_clip": 1.05149341,
      "balance_loss_mlp": 1.02336264,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.05353669642571,
      "language_loss": 0.82407284,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.8460983,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.5689592361450195
    },
    {
      "auxiliary_loss_clip": 0.01058821,
      "auxiliary_loss_mlp": 0.01002072,
      "balance_loss_clip": 1.02322483,
      "balance_loss_mlp": 1.00110626,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7908244683860826,
      "language_loss": 0.57052118,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59113014,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.3502187728881836
    },
    {
      "auxiliary_loss_clip": 0.01169832,
      "auxiliary_loss_mlp": 0.01025537,
      "balance_loss_clip": 1.05204272,
      "balance_loss_mlp": 1.01744831,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.7226521477450778,
      "language_loss": 0.69683594,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.71878958,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.808497667312622
    },
    {
      "auxiliary_loss_clip": 0.01165128,
      "auxiliary_loss_mlp": 0.00901253,
      "balance_loss_clip": 1.05000067,
      "balance_loss_mlp": 1.00086355,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.092668552014996,
      "language_loss": 0.74360847,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.76427233,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.6418728828430176
    },
    {
      "auxiliary_loss_clip": 0.01174502,
      "auxiliary_loss_mlp": 0.00901308,
      "balance_loss_clip": 1.05080605,
      "balance_loss_mlp": 1.00088215,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.33427616135505,
      "language_loss": 0.7450428,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76580089,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 3.5723876953125
    },
    {
      "auxiliary_loss_clip": 0.01113192,
      "auxiliary_loss_mlp": 0.0102412,
      "balance_loss_clip": 1.04011726,
      "balance_loss_mlp": 1.01663637,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.7888221703593772,
      "language_loss": 0.81212604,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83349919,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.7589058876037598
    },
    {
      "auxiliary_loss_clip": 0.01167514,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.05194902,
      "balance_loss_mlp": 1.02003574,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 1.8170544930268226,
      "language_loss": 0.73788232,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75983632,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.6546170711517334
    },
    {
      "auxiliary_loss_clip": 0.01156168,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.04655242,
      "balance_loss_mlp": 1.01679063,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.8240370536048551,
      "language_loss": 0.7009747,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72277975,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.8101284503936768
    },
    {
      "auxiliary_loss_clip": 0.01136989,
      "auxiliary_loss_mlp": 0.01027438,
      "balance_loss_clip": 1.04013562,
      "balance_loss_mlp": 1.01923013,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.6232246726822739,
      "language_loss": 0.83143282,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85307705,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.760885238647461
    },
    {
      "auxiliary_loss_clip": 0.01164913,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.04920518,
      "balance_loss_mlp": 1.01932263,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 1.8884286623678659,
      "language_loss": 0.77657676,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.79849213,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.772399425506592
    },
    {
      "auxiliary_loss_clip": 0.0117595,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.05130267,
      "balance_loss_mlp": 1.01696515,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 1.8776616674742697,
      "language_loss": 0.76330107,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78530383,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.5717146396636963
    },
    {
      "auxiliary_loss_clip": 0.01169219,
      "auxiliary_loss_mlp": 0.00901167,
      "balance_loss_clip": 1.05172992,
      "balance_loss_mlp": 1.00087023,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 2.6844194015491287,
      "language_loss": 0.64824098,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66894484,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.641005516052246
    },
    {
      "auxiliary_loss_clip": 0.01165078,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.05116987,
      "balance_loss_mlp": 1.018296,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.2962215688122933,
      "language_loss": 0.78966093,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81157422,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 3.532404661178589
    },
    {
      "auxiliary_loss_clip": 0.01152782,
      "auxiliary_loss_mlp": 0.00901451,
      "balance_loss_clip": 1.04510903,
      "balance_loss_mlp": 1.00088286,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.9071042075405913,
      "language_loss": 0.77835786,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.79890025,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.728947401046753
    },
    {
      "auxiliary_loss_clip": 0.01133955,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04364252,
      "balance_loss_mlp": 1.01849723,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 2.2290544747980587,
      "language_loss": 0.79931092,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.8209157,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.685652017593384
    },
    {
      "auxiliary_loss_clip": 0.01118131,
      "auxiliary_loss_mlp": 0.0102563,
      "balance_loss_clip": 1.04159224,
      "balance_loss_mlp": 1.0182631,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.6930747425973784,
      "language_loss": 0.76087332,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.78231096,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.825270175933838
    },
    {
      "auxiliary_loss_clip": 0.01158501,
      "auxiliary_loss_mlp": 0.01027811,
      "balance_loss_clip": 1.05003715,
      "balance_loss_mlp": 1.02011871,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 1.8364062361141336,
      "language_loss": 0.72981334,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.75167644,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.6242282390594482
    },
    {
      "auxiliary_loss_clip": 0.01156013,
      "auxiliary_loss_mlp": 0.01028008,
      "balance_loss_clip": 1.04819345,
      "balance_loss_mlp": 1.02053928,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.6573938180795174,
      "language_loss": 0.77315462,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79499477,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.71431565284729
    },
    {
      "auxiliary_loss_clip": 0.0116754,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.05394197,
      "balance_loss_mlp": 1.02062583,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.2000264265787597,
      "language_loss": 0.81731766,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83927506,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.595515727996826
    },
    {
      "auxiliary_loss_clip": 0.01139471,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.04683447,
      "balance_loss_mlp": 1.01966453,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.339245556095941,
      "language_loss": 0.77083635,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79250181,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.6615328788757324
    },
    {
      "auxiliary_loss_clip": 0.01157707,
      "auxiliary_loss_mlp": 0.01018956,
      "balance_loss_clip": 1.04831493,
      "balance_loss_mlp": 1.01187241,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.4825503628621224,
      "language_loss": 0.73549342,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75726008,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.6813836097717285
    },
    {
      "auxiliary_loss_clip": 0.01158824,
      "auxiliary_loss_mlp": 0.01028112,
      "balance_loss_clip": 1.05262589,
      "balance_loss_mlp": 1.02056587,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.592009302850988,
      "language_loss": 0.76645082,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78832018,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 3.661283016204834
    },
    {
      "auxiliary_loss_clip": 0.01170058,
      "auxiliary_loss_mlp": 0.01031972,
      "balance_loss_clip": 1.04866242,
      "balance_loss_mlp": 1.02433634,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 1.9921777746845815,
      "language_loss": 0.70770085,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.72972113,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.67716646194458
    },
    {
      "auxiliary_loss_clip": 0.01145892,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.04503822,
      "balance_loss_mlp": 1.02050745,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.236534103386817,
      "language_loss": 0.89154363,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91328478,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.6617298126220703
    },
    {
      "auxiliary_loss_clip": 0.01163685,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.04689813,
      "balance_loss_mlp": 1.01793182,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.3638815037514593,
      "language_loss": 0.7548216,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77672112,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 3.5690362453460693
    },
    {
      "auxiliary_loss_clip": 0.01164502,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.04796171,
      "balance_loss_mlp": 1.01910865,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.629679927151111,
      "language_loss": 0.73811138,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76002896,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.725435495376587
    },
    {
      "auxiliary_loss_clip": 0.01149919,
      "auxiliary_loss_mlp": 0.01022289,
      "balance_loss_clip": 1.05139911,
      "balance_loss_mlp": 1.01411724,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 1.8270591086078196,
      "language_loss": 0.70963299,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73135507,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.8012447357177734
    },
    {
      "auxiliary_loss_clip": 0.01162223,
      "auxiliary_loss_mlp": 0.01024499,
      "balance_loss_clip": 1.04864514,
      "balance_loss_mlp": 1.01689982,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.865225373686667,
      "language_loss": 0.78010172,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80196893,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.6300652027130127
    },
    {
      "auxiliary_loss_clip": 0.01161741,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.05052209,
      "balance_loss_mlp": 1.02228093,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.7604116707189401,
      "language_loss": 0.78074139,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80265677,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.7035505771636963
    },
    {
      "auxiliary_loss_clip": 0.011768,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.05172801,
      "balance_loss_mlp": 1.0179075,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.838377935109333,
      "language_loss": 0.75155652,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.77357745,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.6551284790039062
    },
    {
      "auxiliary_loss_clip": 0.01170235,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.05236447,
      "balance_loss_mlp": 1.0202533,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 1.9963503235910651,
      "language_loss": 0.78987795,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.81186074,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 3.5774073600769043
    },
    {
      "auxiliary_loss_clip": 0.01167236,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04947805,
      "balance_loss_mlp": 1.01948524,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.7384713187135579,
      "language_loss": 0.76461935,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78655863,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.7061829566955566
    },
    {
      "auxiliary_loss_clip": 0.01153721,
      "auxiliary_loss_mlp": 0.01027826,
      "balance_loss_clip": 1.04922247,
      "balance_loss_mlp": 1.01929688,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.313936429357027,
      "language_loss": 0.8538177,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.87563324,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.9105417728424072
    },
    {
      "auxiliary_loss_clip": 0.0117589,
      "auxiliary_loss_mlp": 0.00901434,
      "balance_loss_clip": 1.0494535,
      "balance_loss_mlp": 1.00072813,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.623711119209509,
      "language_loss": 0.84479308,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86556637,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.5632452964782715
    },
    {
      "auxiliary_loss_clip": 0.01146022,
      "auxiliary_loss_mlp": 0.01023708,
      "balance_loss_clip": 1.04558492,
      "balance_loss_mlp": 1.01594698,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 5.178305090782796,
      "language_loss": 0.80403924,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82573652,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.734869956970215
    },
    {
      "auxiliary_loss_clip": 0.01151942,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.0511198,
      "balance_loss_mlp": 1.02300382,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 1.9180801386089927,
      "language_loss": 0.83350968,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85533518,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.6470351219177246
    },
    {
      "auxiliary_loss_clip": 0.01162318,
      "auxiliary_loss_mlp": 0.0102568,
      "balance_loss_clip": 1.04947746,
      "balance_loss_mlp": 1.01805615,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 1.687631675216991,
      "language_loss": 0.76704288,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78892285,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.736783981323242
    },
    {
      "auxiliary_loss_clip": 0.0107827,
      "auxiliary_loss_mlp": 0.01001226,
      "balance_loss_clip": 1.01832533,
      "balance_loss_mlp": 1.00034416,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8412557115647084,
      "language_loss": 0.57063746,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59143245,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.1395394802093506
    },
    {
      "auxiliary_loss_clip": 0.01168103,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.05230308,
      "balance_loss_mlp": 1.01993632,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 2.1917821870140313,
      "language_loss": 0.7559973,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77795124,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.671891450881958
    },
    {
      "auxiliary_loss_clip": 0.01146409,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04481494,
      "balance_loss_mlp": 1.01756036,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.7008345013110708,
      "language_loss": 0.83319914,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.85491133,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 3.643240213394165
    },
    {
      "auxiliary_loss_clip": 0.01158659,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.05053639,
      "balance_loss_mlp": 1.01848269,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.261746445010235,
      "language_loss": 0.82641816,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84826291,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.6571884155273438
    },
    {
      "auxiliary_loss_clip": 0.0115873,
      "auxiliary_loss_mlp": 0.01025204,
      "balance_loss_clip": 1.04835796,
      "balance_loss_mlp": 1.01797128,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 2.6607824932610664,
      "language_loss": 0.90212888,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.9239682,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 2.6455211639404297
    },
    {
      "auxiliary_loss_clip": 0.01147471,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.04795742,
      "balance_loss_mlp": 1.02146006,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.804918835542083,
      "language_loss": 0.83145857,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85322344,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.696467161178589
    },
    {
      "auxiliary_loss_clip": 0.01164222,
      "auxiliary_loss_mlp": 0.01024272,
      "balance_loss_clip": 1.05099511,
      "balance_loss_mlp": 1.01716399,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 2.665092181525125,
      "language_loss": 0.75350386,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77538884,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.6559321880340576
    },
    {
      "auxiliary_loss_clip": 0.01161248,
      "auxiliary_loss_mlp": 0.01024646,
      "balance_loss_clip": 1.05094528,
      "balance_loss_mlp": 1.01757348,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 3.45546291044336,
      "language_loss": 0.72582114,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74768007,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6114041805267334
    },
    {
      "auxiliary_loss_clip": 0.01163937,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.04877353,
      "balance_loss_mlp": 1.01699638,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.2732898449147303,
      "language_loss": 0.78346109,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.80534923,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.591301202774048
    },
    {
      "auxiliary_loss_clip": 0.01176186,
      "auxiliary_loss_mlp": 0.01030705,
      "balance_loss_clip": 1.05332482,
      "balance_loss_mlp": 1.02322471,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.271118595297957,
      "language_loss": 0.75743902,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.77950788,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.547285795211792
    },
    {
      "auxiliary_loss_clip": 0.01076511,
      "auxiliary_loss_mlp": 0.01004134,
      "balance_loss_clip": 1.01829147,
      "balance_loss_mlp": 1.00331783,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7511450240456101,
      "language_loss": 0.57878357,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59959006,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.036863327026367
    },
    {
      "auxiliary_loss_clip": 0.01155313,
      "auxiliary_loss_mlp": 0.01028632,
      "balance_loss_clip": 1.04757261,
      "balance_loss_mlp": 1.02117562,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.656716261996889,
      "language_loss": 0.6410352,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.6628747,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 3.711679458618164
    },
    {
      "auxiliary_loss_clip": 0.01175223,
      "auxiliary_loss_mlp": 0.01025672,
      "balance_loss_clip": 1.05247188,
      "balance_loss_mlp": 1.01888275,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 1.9260640058588723,
      "language_loss": 0.71769285,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.73970187,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.5659823417663574
    },
    {
      "auxiliary_loss_clip": 0.01141678,
      "auxiliary_loss_mlp": 0.00900991,
      "balance_loss_clip": 1.05259323,
      "balance_loss_mlp": 1.0008359,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.719950789126842,
      "language_loss": 0.78701103,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.80743772,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.782705545425415
    },
    {
      "auxiliary_loss_clip": 0.01152195,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.04799509,
      "balance_loss_mlp": 1.02099752,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 2.1789846013723,
      "language_loss": 0.82142407,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84323466,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 3.7467448711395264
    },
    {
      "auxiliary_loss_clip": 0.01167909,
      "auxiliary_loss_mlp": 0.01025079,
      "balance_loss_clip": 1.05091763,
      "balance_loss_mlp": 1.0175631,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.252650529859018,
      "language_loss": 0.84150106,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86343098,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.561061143875122
    },
    {
      "auxiliary_loss_clip": 0.01164163,
      "auxiliary_loss_mlp": 0.01025338,
      "balance_loss_clip": 1.05145943,
      "balance_loss_mlp": 1.01768458,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 2.270938213929374,
      "language_loss": 0.71377194,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73566693,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.6192691326141357
    },
    {
      "auxiliary_loss_clip": 0.01166257,
      "auxiliary_loss_mlp": 0.01021492,
      "balance_loss_clip": 1.05090952,
      "balance_loss_mlp": 1.01391625,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.9369454352369637,
      "language_loss": 0.8171525,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.83903003,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.651810884475708
    },
    {
      "auxiliary_loss_clip": 0.01165557,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.05010366,
      "balance_loss_mlp": 1.01780164,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.7321030528916403,
      "language_loss": 0.81178385,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.8336885,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.5621464252471924
    },
    {
      "auxiliary_loss_clip": 0.01162607,
      "auxiliary_loss_mlp": 0.01028606,
      "balance_loss_clip": 1.05115736,
      "balance_loss_mlp": 1.0214479,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 1.7275584132566641,
      "language_loss": 0.76374102,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78565323,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.635098934173584
    },
    {
      "auxiliary_loss_clip": 0.01172955,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05064261,
      "balance_loss_mlp": 1.01961827,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.768606476480276,
      "language_loss": 0.7434234,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76542377,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 3.522676944732666
    },
    {
      "auxiliary_loss_clip": 0.01169204,
      "auxiliary_loss_mlp": 0.01023239,
      "balance_loss_clip": 1.05306935,
      "balance_loss_mlp": 1.01586342,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 2.0213983283699517,
      "language_loss": 0.69978583,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72171032,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.6456503868103027
    },
    {
      "auxiliary_loss_clip": 0.01168799,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.05307591,
      "balance_loss_mlp": 1.02233887,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 2.17717180495402,
      "language_loss": 0.83415264,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85613644,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.699320077896118
    },
    {
      "auxiliary_loss_clip": 0.01149624,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.04587424,
      "balance_loss_mlp": 1.01853395,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.758872722617398,
      "language_loss": 0.75574774,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.77750492,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.7627909183502197
    },
    {
      "auxiliary_loss_clip": 0.01171275,
      "auxiliary_loss_mlp": 0.01024059,
      "balance_loss_clip": 1.05329013,
      "balance_loss_mlp": 1.01681113,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.853016454243577,
      "language_loss": 0.78174907,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80370235,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.697962760925293
    },
    {
      "auxiliary_loss_clip": 0.0118161,
      "auxiliary_loss_mlp": 0.01027556,
      "balance_loss_clip": 1.05361485,
      "balance_loss_mlp": 1.01966465,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 2.280830947619484,
      "language_loss": 0.6736747,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.69576639,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.6355926990509033
    },
    {
      "auxiliary_loss_clip": 0.01154627,
      "auxiliary_loss_mlp": 0.01027396,
      "balance_loss_clip": 1.0503267,
      "balance_loss_mlp": 1.01973033,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 1.8911063314469378,
      "language_loss": 0.80640882,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.82822907,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.7213850021362305
    },
    {
      "auxiliary_loss_clip": 0.01065644,
      "auxiliary_loss_mlp": 0.01001465,
      "balance_loss_clip": 1.01974869,
      "balance_loss_mlp": 1.00055873,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7355774106675421,
      "language_loss": 0.54276824,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56343937,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.2211825847625732
    },
    {
      "auxiliary_loss_clip": 0.0115848,
      "auxiliary_loss_mlp": 0.01026908,
      "balance_loss_clip": 1.0503335,
      "balance_loss_mlp": 1.01940405,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.740215087259378,
      "language_loss": 0.841102,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.86295587,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 3.5918524265289307
    },
    {
      "auxiliary_loss_clip": 0.0114876,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04788947,
      "balance_loss_mlp": 1.01902688,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 2.528232164254631,
      "language_loss": 0.72015285,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.7419107,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.639106512069702
    },
    {
      "auxiliary_loss_clip": 0.01159829,
      "auxiliary_loss_mlp": 0.01024211,
      "balance_loss_clip": 1.04663193,
      "balance_loss_mlp": 1.0167098,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 2.080036510767015,
      "language_loss": 0.75641996,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77826035,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.5962178707122803
    },
    {
      "auxiliary_loss_clip": 0.01146254,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 1.04736555,
      "balance_loss_mlp": 1.01869142,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.0124900815675693,
      "language_loss": 0.71143341,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73315942,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 2.7046520709991455
    },
    {
      "auxiliary_loss_clip": 0.01168395,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.04995811,
      "balance_loss_mlp": 1.02240884,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.535204304476419,
      "language_loss": 0.76026535,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78224069,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.621936321258545
    },
    {
      "auxiliary_loss_clip": 0.0105742,
      "auxiliary_loss_mlp": 0.0100233,
      "balance_loss_clip": 1.02308154,
      "balance_loss_mlp": 1.00143552,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8200557615929751,
      "language_loss": 0.58938897,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.60998642,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.2776036262512207
    },
    {
      "auxiliary_loss_clip": 0.01154953,
      "auxiliary_loss_mlp": 0.01028528,
      "balance_loss_clip": 1.04605222,
      "balance_loss_mlp": 1.02067804,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 3.466073662201616,
      "language_loss": 0.70432937,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72616416,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 3.1195755004882812
    },
    {
      "auxiliary_loss_clip": 0.01143592,
      "auxiliary_loss_mlp": 0.0102446,
      "balance_loss_clip": 1.04280519,
      "balance_loss_mlp": 1.01672935,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.8813552954892372,
      "language_loss": 0.76830661,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.78998715,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.7446625232696533
    },
    {
      "auxiliary_loss_clip": 0.01162821,
      "auxiliary_loss_mlp": 0.01027951,
      "balance_loss_clip": 1.04803848,
      "balance_loss_mlp": 1.0207715,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.7741991994236208,
      "language_loss": 0.76557541,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78748316,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.7136385440826416
    },
    {
      "auxiliary_loss_clip": 0.01166391,
      "auxiliary_loss_mlp": 0.0102244,
      "balance_loss_clip": 1.04925585,
      "balance_loss_mlp": 1.01482868,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.6271536052490563,
      "language_loss": 0.70795459,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.7298429,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 3.7938036918640137
    },
    {
      "auxiliary_loss_clip": 0.01162472,
      "auxiliary_loss_mlp": 0.01023779,
      "balance_loss_clip": 1.05105257,
      "balance_loss_mlp": 1.01626301,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 2.038311166519366,
      "language_loss": 0.73281455,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75467706,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.698824882507324
    },
    {
      "auxiliary_loss_clip": 0.01177051,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.05249417,
      "balance_loss_mlp": 1.02066517,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.9583840809333561,
      "language_loss": 0.77459931,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.79665446,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.7036197185516357
    },
    {
      "auxiliary_loss_clip": 0.0115166,
      "auxiliary_loss_mlp": 0.01026252,
      "balance_loss_clip": 1.04439116,
      "balance_loss_mlp": 1.01846218,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 3.106649597385068,
      "language_loss": 0.75346267,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77524173,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 3.5737476348876953
    },
    {
      "auxiliary_loss_clip": 0.01166463,
      "auxiliary_loss_mlp": 0.01023554,
      "balance_loss_clip": 1.05061257,
      "balance_loss_mlp": 1.01594877,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 3.535434805110731,
      "language_loss": 0.70157552,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72347569,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.659900188446045
    },
    {
      "auxiliary_loss_clip": 0.01170347,
      "auxiliary_loss_mlp": 0.01022521,
      "balance_loss_clip": 1.05278206,
      "balance_loss_mlp": 1.01478446,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 15.718644815924922,
      "language_loss": 0.8118307,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83375931,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.6210405826568604
    },
    {
      "auxiliary_loss_clip": 0.01079233,
      "auxiliary_loss_mlp": 0.01014656,
      "balance_loss_clip": 1.03249586,
      "balance_loss_mlp": 1.01359546,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7892384416684509,
      "language_loss": 0.53354776,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55448663,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.2148385047912598
    },
    {
      "auxiliary_loss_clip": 0.01172217,
      "auxiliary_loss_mlp": 0.01025706,
      "balance_loss_clip": 1.0524807,
      "balance_loss_mlp": 1.01815963,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 1.87009672354054,
      "language_loss": 0.76559168,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.78757095,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.6019110679626465
    },
    {
      "auxiliary_loss_clip": 0.01159729,
      "auxiliary_loss_mlp": 0.01024063,
      "balance_loss_clip": 1.05200291,
      "balance_loss_mlp": 1.01596856,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 2.34380168762278,
      "language_loss": 0.7608875,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78272545,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.627584457397461
    },
    {
      "auxiliary_loss_clip": 0.01176573,
      "auxiliary_loss_mlp": 0.00900979,
      "balance_loss_clip": 1.05210876,
      "balance_loss_mlp": 1.00081778,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.650310598935144,
      "language_loss": 0.71833706,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.73911256,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.6733062267303467
    },
    {
      "auxiliary_loss_clip": 0.01138343,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 1.04252613,
      "balance_loss_mlp": 1.01751447,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 2.719193489285753,
      "language_loss": 0.63075173,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65238655,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 3.655977249145508
    },
    {
      "auxiliary_loss_clip": 0.01151892,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 1.05490279,
      "balance_loss_mlp": 1.01946175,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.6179626196023962,
      "language_loss": 0.74083221,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76262379,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.724538803100586
    },
    {
      "auxiliary_loss_clip": 0.01158708,
      "auxiliary_loss_mlp": 0.01023452,
      "balance_loss_clip": 1.05178428,
      "balance_loss_mlp": 1.01571274,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 2.0580851614743803,
      "language_loss": 0.61973596,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64155751,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.6859591007232666
    },
    {
      "auxiliary_loss_clip": 0.01178616,
      "auxiliary_loss_mlp": 0.01032181,
      "balance_loss_clip": 1.05254614,
      "balance_loss_mlp": 1.02409601,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 2.6908985460145276,
      "language_loss": 0.82099187,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.84309983,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.6438581943511963
    },
    {
      "auxiliary_loss_clip": 0.01174351,
      "auxiliary_loss_mlp": 0.01026909,
      "balance_loss_clip": 1.05069947,
      "balance_loss_mlp": 1.01965523,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 2.0517190098841933,
      "language_loss": 0.79225856,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81427121,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.691185235977173
    },
    {
      "auxiliary_loss_clip": 0.01161308,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.04745781,
      "balance_loss_mlp": 1.01926053,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 3.040757692276165,
      "language_loss": 0.77456677,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.79645097,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.648895502090454
    },
    {
      "auxiliary_loss_clip": 0.01158996,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.0452261,
      "balance_loss_mlp": 1.02210593,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 2.788672025274508,
      "language_loss": 0.726457,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74834353,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.607494831085205
    },
    {
      "auxiliary_loss_clip": 0.01159372,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.04790473,
      "balance_loss_mlp": 1.02063823,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 1.7714671985912256,
      "language_loss": 0.7611798,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78305751,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 3.54215931892395
    },
    {
      "auxiliary_loss_clip": 0.01168324,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 1.0519886,
      "balance_loss_mlp": 1.01917553,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 2.07976872674913,
      "language_loss": 0.75991881,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.7818678,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.6216373443603516
    },
    {
      "auxiliary_loss_clip": 0.01145355,
      "auxiliary_loss_mlp": 0.01023118,
      "balance_loss_clip": 1.04645705,
      "balance_loss_mlp": 1.01556957,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 1.98539770088757,
      "language_loss": 0.70699936,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72868413,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.6334216594696045
    },
    {
      "auxiliary_loss_clip": 0.01161929,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.04938686,
      "balance_loss_mlp": 1.01815486,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 1.96613666877689,
      "language_loss": 0.88036615,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90224886,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.657768726348877
    },
    {
      "auxiliary_loss_clip": 0.01174814,
      "auxiliary_loss_mlp": 0.01022404,
      "balance_loss_clip": 1.04925835,
      "balance_loss_mlp": 1.01559687,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.810962149478429,
      "language_loss": 0.77330041,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79527259,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 2.545969009399414
    },
    {
      "auxiliary_loss_clip": 0.01137819,
      "auxiliary_loss_mlp": 0.00901426,
      "balance_loss_clip": 1.04539084,
      "balance_loss_mlp": 1.00078022,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.0820601431420793,
      "language_loss": 0.7617054,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78209782,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.6733310222625732
    },
    {
      "auxiliary_loss_clip": 0.01170774,
      "auxiliary_loss_mlp": 0.01021959,
      "balance_loss_clip": 1.05103755,
      "balance_loss_mlp": 1.01482773,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 2.1151465905743794,
      "language_loss": 0.74473548,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76666284,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.735957384109497
    },
    {
      "auxiliary_loss_clip": 0.01168719,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.05239511,
      "balance_loss_mlp": 1.02208579,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.4873665614732223,
      "language_loss": 0.63741571,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.65939736,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.6458470821380615
    },
    {
      "auxiliary_loss_clip": 0.01160852,
      "auxiliary_loss_mlp": 0.01019962,
      "balance_loss_clip": 1.04765224,
      "balance_loss_mlp": 1.01189756,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 2.305988934746881,
      "language_loss": 0.72395289,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.74576104,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.6141343116760254
    },
    {
      "auxiliary_loss_clip": 0.01176232,
      "auxiliary_loss_mlp": 0.0102186,
      "balance_loss_clip": 1.05048895,
      "balance_loss_mlp": 1.01444173,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.2983213737333306,
      "language_loss": 0.63813472,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.66011566,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.504434585571289
    },
    {
      "auxiliary_loss_clip": 0.0116974,
      "auxiliary_loss_mlp": 0.00900404,
      "balance_loss_clip": 1.05047786,
      "balance_loss_mlp": 1.00063014,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 3.1966889040324244,
      "language_loss": 0.7844547,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.80515611,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 3.608210325241089
    },
    {
      "auxiliary_loss_clip": 0.01156862,
      "auxiliary_loss_mlp": 0.01026453,
      "balance_loss_clip": 1.04916811,
      "balance_loss_mlp": 1.01948202,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.8182475038224752,
      "language_loss": 0.81411821,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83595133,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.680447578430176
    },
    {
      "auxiliary_loss_clip": 0.01162017,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.04808497,
      "balance_loss_mlp": 1.0197885,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 2.555513048176728,
      "language_loss": 0.72999597,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75188947,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.6762704849243164
    },
    {
      "auxiliary_loss_clip": 0.01144375,
      "auxiliary_loss_mlp": 0.01027741,
      "balance_loss_clip": 1.04450572,
      "balance_loss_mlp": 1.02026057,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.8634665264715222,
      "language_loss": 0.81458384,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83630502,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 3.725374460220337
    },
    {
      "auxiliary_loss_clip": 0.01163743,
      "auxiliary_loss_mlp": 0.01022773,
      "balance_loss_clip": 1.04963136,
      "balance_loss_mlp": 1.01523304,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 5.6803812918436085,
      "language_loss": 0.77473539,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79660058,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.689065933227539
    },
    {
      "auxiliary_loss_clip": 0.01152544,
      "auxiliary_loss_mlp": 0.0102274,
      "balance_loss_clip": 1.04840517,
      "balance_loss_mlp": 1.01462793,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 2.0828163820004364,
      "language_loss": 0.73974842,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76150125,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.7455081939697266
    },
    {
      "auxiliary_loss_clip": 0.01171499,
      "auxiliary_loss_mlp": 0.01029056,
      "balance_loss_clip": 1.04867196,
      "balance_loss_mlp": 1.02092004,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 2.427967282096344,
      "language_loss": 0.72834587,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.75035143,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.642368793487549
    },
    {
      "auxiliary_loss_clip": 0.01137796,
      "auxiliary_loss_mlp": 0.01026026,
      "balance_loss_clip": 1.04259968,
      "balance_loss_mlp": 1.01820016,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.190984570066239,
      "language_loss": 0.61667877,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.63831699,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.9895801544189453
    },
    {
      "auxiliary_loss_clip": 0.01166006,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 1.05017555,
      "balance_loss_mlp": 1.01752806,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 1.6014740786689698,
      "language_loss": 0.73026741,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75217831,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.6478753089904785
    },
    {
      "auxiliary_loss_clip": 0.01176674,
      "auxiliary_loss_mlp": 0.01023469,
      "balance_loss_clip": 1.0506835,
      "balance_loss_mlp": 1.0158484,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 1.8856145160815942,
      "language_loss": 0.7242099,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74621129,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 3.622906446456909
    },
    {
      "auxiliary_loss_clip": 0.01165152,
      "auxiliary_loss_mlp": 0.01023405,
      "balance_loss_clip": 1.05154777,
      "balance_loss_mlp": 1.01602006,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 2.147782259681412,
      "language_loss": 0.68925107,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71113658,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.644580364227295
    },
    {
      "auxiliary_loss_clip": 0.01118649,
      "auxiliary_loss_mlp": 0.01024025,
      "balance_loss_clip": 1.04115009,
      "balance_loss_mlp": 1.01636636,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 2.429142565717317,
      "language_loss": 0.84706604,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.86849278,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.751391887664795
    },
    {
      "auxiliary_loss_clip": 0.01180784,
      "auxiliary_loss_mlp": 0.01031122,
      "balance_loss_clip": 1.05427706,
      "balance_loss_mlp": 1.02297997,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 2.32012694184744,
      "language_loss": 0.82298833,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84510744,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.6101441383361816
    },
    {
      "auxiliary_loss_clip": 0.01147912,
      "auxiliary_loss_mlp": 0.01022443,
      "balance_loss_clip": 1.04798436,
      "balance_loss_mlp": 1.01493239,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.7045604311409848,
      "language_loss": 0.7295745,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75127798,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.6615374088287354
    },
    {
      "auxiliary_loss_clip": 0.01136964,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.04376292,
      "balance_loss_mlp": 1.01928294,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.9503421612693008,
      "language_loss": 0.69091612,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71255314,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.6946375370025635
    },
    {
      "auxiliary_loss_clip": 0.01144433,
      "auxiliary_loss_mlp": 0.01021895,
      "balance_loss_clip": 1.04815185,
      "balance_loss_mlp": 1.01433146,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 2.430379687226768,
      "language_loss": 0.7995131,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82117635,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.6700260639190674
    },
    {
      "auxiliary_loss_clip": 0.01161445,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.04933834,
      "balance_loss_mlp": 1.02127337,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.234587057219957,
      "language_loss": 0.78601527,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80792969,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.712355852127075
    },
    {
      "auxiliary_loss_clip": 0.01147209,
      "auxiliary_loss_mlp": 0.01024584,
      "balance_loss_clip": 1.04324853,
      "balance_loss_mlp": 1.0169127,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 1.7758159561381606,
      "language_loss": 0.75756168,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77927971,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 3.611006259918213
    },
    {
      "auxiliary_loss_clip": 0.01168608,
      "auxiliary_loss_mlp": 0.01027896,
      "balance_loss_clip": 1.05251026,
      "balance_loss_mlp": 1.02096999,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 1.9716821968187768,
      "language_loss": 0.83262098,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85458601,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.677927017211914
    },
    {
      "auxiliary_loss_clip": 0.0116789,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.05110919,
      "balance_loss_mlp": 1.01882493,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.488763265477319,
      "language_loss": 0.73763126,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75957441,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.690011739730835
    },
    {
      "auxiliary_loss_clip": 0.0115163,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.04595125,
      "balance_loss_mlp": 1.01781249,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 2.074924188435479,
      "language_loss": 0.80069637,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82246864,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.77567458152771
    },
    {
      "auxiliary_loss_clip": 0.01160321,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.04805601,
      "balance_loss_mlp": 1.01835966,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.020547048138393,
      "language_loss": 0.85530019,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87716848,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.628356456756592
    },
    {
      "auxiliary_loss_clip": 0.01145886,
      "auxiliary_loss_mlp": 0.01023813,
      "balance_loss_clip": 1.045784,
      "balance_loss_mlp": 1.01648808,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.6810286056560897,
      "language_loss": 0.7885114,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.81020844,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.711009979248047
    },
    {
      "auxiliary_loss_clip": 0.01144872,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04327977,
      "balance_loss_mlp": 1.01832867,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 3.7149457210504297,
      "language_loss": 0.79296476,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81467712,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.6976211071014404
    },
    {
      "auxiliary_loss_clip": 0.01165744,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.05005217,
      "balance_loss_mlp": 1.01834071,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 2.623664287316909,
      "language_loss": 0.823946,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84586048,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.574394702911377
    },
    {
      "auxiliary_loss_clip": 0.01133463,
      "auxiliary_loss_mlp": 0.00901146,
      "balance_loss_clip": 1.04537606,
      "balance_loss_mlp": 1.00094855,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.8646264058290631,
      "language_loss": 0.78086931,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.80121541,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.775176525115967
    },
    {
      "auxiliary_loss_clip": 0.01126442,
      "auxiliary_loss_mlp": 0.00901368,
      "balance_loss_clip": 1.03905189,
      "balance_loss_mlp": 1.00091326,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.0380518719450245,
      "language_loss": 0.63536251,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65564066,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.753894805908203
    },
    {
      "auxiliary_loss_clip": 0.01154586,
      "auxiliary_loss_mlp": 0.01021666,
      "balance_loss_clip": 1.04776895,
      "balance_loss_mlp": 1.0146029,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.6666081816282066,
      "language_loss": 0.67929053,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70105302,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 3.53601336479187
    },
    {
      "auxiliary_loss_clip": 0.01145154,
      "auxiliary_loss_mlp": 0.01022765,
      "balance_loss_clip": 1.04415476,
      "balance_loss_mlp": 1.01545167,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 3.507828528823625,
      "language_loss": 0.60229623,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.6239754,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.8449082374572754
    },
    {
      "auxiliary_loss_clip": 0.01165289,
      "auxiliary_loss_mlp": 0.01031626,
      "balance_loss_clip": 1.05054104,
      "balance_loss_mlp": 1.02399635,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 2.924643187287513,
      "language_loss": 0.75181788,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77378702,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.5800588130950928
    },
    {
      "auxiliary_loss_clip": 0.01131255,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.04422259,
      "balance_loss_mlp": 1.01732659,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.720517602047458,
      "language_loss": 0.75332332,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77488244,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 3.724824905395508
    },
    {
      "auxiliary_loss_clip": 0.01129698,
      "auxiliary_loss_mlp": 0.0103349,
      "balance_loss_clip": 1.04712939,
      "balance_loss_mlp": 1.02528846,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.1638338768891705,
      "language_loss": 0.81441337,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.83604527,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.838578939437866
    },
    {
      "auxiliary_loss_clip": 0.01153499,
      "auxiliary_loss_mlp": 0.00900594,
      "balance_loss_clip": 1.0454998,
      "balance_loss_mlp": 1.0009129,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.0003640284090887,
      "language_loss": 0.78540039,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80594134,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6292765140533447
    },
    {
      "auxiliary_loss_clip": 0.01158985,
      "auxiliary_loss_mlp": 0.01023278,
      "balance_loss_clip": 1.04823995,
      "balance_loss_mlp": 1.01567209,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.5945905157474118,
      "language_loss": 0.79980731,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82163,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.6589176654815674
    },
    {
      "auxiliary_loss_clip": 0.01166891,
      "auxiliary_loss_mlp": 0.0101988,
      "balance_loss_clip": 1.05084491,
      "balance_loss_mlp": 1.01298392,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 2.1542423361978043,
      "language_loss": 0.73535657,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.7572242,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.6045241355895996
    },
    {
      "auxiliary_loss_clip": 0.01169161,
      "auxiliary_loss_mlp": 0.01023299,
      "balance_loss_clip": 1.05217147,
      "balance_loss_mlp": 1.01601791,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 2.013903066586644,
      "language_loss": 0.77226794,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79419255,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.655245542526245
    },
    {
      "auxiliary_loss_clip": 0.01167134,
      "auxiliary_loss_mlp": 0.00901262,
      "balance_loss_clip": 1.04846597,
      "balance_loss_mlp": 1.00101089,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 1.8219731418795087,
      "language_loss": 0.74941647,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.77010041,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 3.5184338092803955
    },
    {
      "auxiliary_loss_clip": 0.0115776,
      "auxiliary_loss_mlp": 0.01022636,
      "balance_loss_clip": 1.04395413,
      "balance_loss_mlp": 1.01544213,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 1.9682211653073944,
      "language_loss": 0.82472229,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84652627,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.6543428897857666
    },
    {
      "auxiliary_loss_clip": 0.01168015,
      "auxiliary_loss_mlp": 0.01026051,
      "balance_loss_clip": 1.05286455,
      "balance_loss_mlp": 1.01819181,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.3855031854708817,
      "language_loss": 0.78174138,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80368209,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.6955559253692627
    },
    {
      "auxiliary_loss_clip": 0.01169829,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 1.05082095,
      "balance_loss_mlp": 1.01764393,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.5027242522655353,
      "language_loss": 0.75350094,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77545607,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.587829113006592
    },
    {
      "auxiliary_loss_clip": 0.01066823,
      "auxiliary_loss_mlp": 0.01004297,
      "balance_loss_clip": 1.02141142,
      "balance_loss_mlp": 1.00333738,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7260924257861595,
      "language_loss": 0.56281686,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58352804,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.2218103408813477
    },
    {
      "auxiliary_loss_clip": 0.0115906,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.04869604,
      "balance_loss_mlp": 1.01845133,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.531840915733041,
      "language_loss": 0.64558077,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66743445,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.6929402351379395
    },
    {
      "auxiliary_loss_clip": 0.01156341,
      "auxiliary_loss_mlp": 0.0090102,
      "balance_loss_clip": 1.05088449,
      "balance_loss_mlp": 1.00081456,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 1.9879704462011831,
      "language_loss": 0.76936442,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.78993797,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.632888078689575
    },
    {
      "auxiliary_loss_clip": 0.01077492,
      "auxiliary_loss_mlp": 0.01001784,
      "balance_loss_clip": 1.02347589,
      "balance_loss_mlp": 1.0008601,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8147350516912031,
      "language_loss": 0.57621217,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59700489,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.169947624206543
    },
    {
      "auxiliary_loss_clip": 0.01159087,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.04829359,
      "balance_loss_mlp": 1.01749623,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.9010422531117532,
      "language_loss": 0.90278018,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92462325,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 3.642531394958496
    },
    {
      "auxiliary_loss_clip": 0.01169495,
      "auxiliary_loss_mlp": 0.01021595,
      "balance_loss_clip": 1.04989266,
      "balance_loss_mlp": 1.01403069,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 2.074071083436737,
      "language_loss": 0.65837181,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68028259,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.7019612789154053
    },
    {
      "auxiliary_loss_clip": 0.01169198,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.05078435,
      "balance_loss_mlp": 1.02107036,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.258852734278703,
      "language_loss": 0.72273314,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74471301,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.650000810623169
    },
    {
      "auxiliary_loss_clip": 0.01160388,
      "auxiliary_loss_mlp": 0.01022949,
      "balance_loss_clip": 1.04702234,
      "balance_loss_mlp": 1.01552808,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.7898224042774062,
      "language_loss": 0.75331628,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77514964,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.643812894821167
    },
    {
      "auxiliary_loss_clip": 0.01144811,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.04855919,
      "balance_loss_mlp": 1.0165869,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.5410608030088362,
      "language_loss": 0.669312,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69100106,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.7726516723632812
    },
    {
      "auxiliary_loss_clip": 0.01157273,
      "auxiliary_loss_mlp": 0.0102723,
      "balance_loss_clip": 1.04952502,
      "balance_loss_mlp": 1.01954091,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.267899977900741,
      "language_loss": 0.80528432,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82712936,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.6767618656158447
    },
    {
      "auxiliary_loss_clip": 0.01176522,
      "auxiliary_loss_mlp": 0.01023327,
      "balance_loss_clip": 1.05225015,
      "balance_loss_mlp": 1.01619554,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 1.887653367311187,
      "language_loss": 0.71015084,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73214936,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.5365936756134033
    },
    {
      "auxiliary_loss_clip": 0.01144183,
      "auxiliary_loss_mlp": 0.01033345,
      "balance_loss_clip": 1.04701042,
      "balance_loss_mlp": 1.02573061,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 2.2906986722185754,
      "language_loss": 0.7049337,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72670901,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.717008352279663
    },
    {
      "auxiliary_loss_clip": 0.01166742,
      "auxiliary_loss_mlp": 0.00901218,
      "balance_loss_clip": 1.04752493,
      "balance_loss_mlp": 1.00072026,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 2.170846145589118,
      "language_loss": 0.79896969,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.81964928,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.6026060581207275
    },
    {
      "auxiliary_loss_clip": 0.01147604,
      "auxiliary_loss_mlp": 0.01025449,
      "balance_loss_clip": 1.04553998,
      "balance_loss_mlp": 1.01768827,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.9877814698398613,
      "language_loss": 0.84406877,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86579919,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 3.6787374019622803
    },
    {
      "auxiliary_loss_clip": 0.01151956,
      "auxiliary_loss_mlp": 0.0103162,
      "balance_loss_clip": 1.04792154,
      "balance_loss_mlp": 1.02340019,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.7761708436834645,
      "language_loss": 0.7067399,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72857571,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.691690444946289
    },
    {
      "auxiliary_loss_clip": 0.01168293,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.05063593,
      "balance_loss_mlp": 1.02085972,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 1.923169651426749,
      "language_loss": 0.69292319,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71489125,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.615493059158325
    },
    {
      "auxiliary_loss_clip": 0.0116239,
      "auxiliary_loss_mlp": 0.01023694,
      "balance_loss_clip": 1.04930449,
      "balance_loss_mlp": 1.01605225,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 2.1137333249113666,
      "language_loss": 0.73677671,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75863755,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.700258255004883
    },
    {
      "auxiliary_loss_clip": 0.01177643,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.0526042,
      "balance_loss_mlp": 1.01957846,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.4018911797745637,
      "language_loss": 0.71734321,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73939484,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 3.521955966949463
    },
    {
      "auxiliary_loss_clip": 0.01161397,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.04687428,
      "balance_loss_mlp": 1.01952469,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 1.947011114401878,
      "language_loss": 0.77982342,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80170816,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.7054243087768555
    },
    {
      "auxiliary_loss_clip": 0.01149999,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.04249716,
      "balance_loss_mlp": 1.01768327,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.9195944899796262,
      "language_loss": 0.76879632,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.79055643,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.668010950088501
    },
    {
      "auxiliary_loss_clip": 0.01153512,
      "auxiliary_loss_mlp": 0.01026864,
      "balance_loss_clip": 1.04816318,
      "balance_loss_mlp": 1.01976776,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 1.858458089695214,
      "language_loss": 0.76877695,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.79058075,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.7106916904449463
    },
    {
      "auxiliary_loss_clip": 0.01132691,
      "auxiliary_loss_mlp": 0.01026454,
      "balance_loss_clip": 1.0437932,
      "balance_loss_mlp": 1.01930773,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 2.140094106160611,
      "language_loss": 0.81998998,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.8415814,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.664357900619507
    },
    {
      "auxiliary_loss_clip": 0.01161877,
      "auxiliary_loss_mlp": 0.01026247,
      "balance_loss_clip": 1.04989707,
      "balance_loss_mlp": 1.01883173,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 2.0354032921284175,
      "language_loss": 0.80424899,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82613027,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.648087978363037
    },
    {
      "auxiliary_loss_clip": 0.01167984,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.0486114,
      "balance_loss_mlp": 1.02222013,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.6393163082277986,
      "language_loss": 0.65954834,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68152308,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 3.623656749725342
    },
    {
      "auxiliary_loss_clip": 0.01144909,
      "auxiliary_loss_mlp": 0.01023872,
      "balance_loss_clip": 1.04204345,
      "balance_loss_mlp": 1.01609969,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 2.4619192147391846,
      "language_loss": 0.75999385,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.78168166,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.7664334774017334
    },
    {
      "auxiliary_loss_clip": 0.01175673,
      "auxiliary_loss_mlp": 0.01023316,
      "balance_loss_clip": 1.04937077,
      "balance_loss_mlp": 1.01619625,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 2.0067608648505924,
      "language_loss": 0.7288577,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.75084758,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.6003661155700684
    },
    {
      "auxiliary_loss_clip": 0.01165608,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.05121636,
      "balance_loss_mlp": 1.02146339,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 2.318133990607548,
      "language_loss": 0.76367372,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78561473,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.622390031814575
    },
    {
      "auxiliary_loss_clip": 0.01168809,
      "auxiliary_loss_mlp": 0.01023317,
      "balance_loss_clip": 1.04824162,
      "balance_loss_mlp": 1.01625681,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.462696459697698,
      "language_loss": 0.79950494,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82142615,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.6021029949188232
    },
    {
      "auxiliary_loss_clip": 0.01119179,
      "auxiliary_loss_mlp": 0.01026682,
      "balance_loss_clip": 1.04397476,
      "balance_loss_mlp": 1.01877284,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 1.922468019606158,
      "language_loss": 0.73216522,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75362384,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.720931053161621
    },
    {
      "auxiliary_loss_clip": 0.01138847,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.04096389,
      "balance_loss_mlp": 1.01899624,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 1.8401488952069143,
      "language_loss": 0.82328016,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84493518,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.663382053375244
    },
    {
      "auxiliary_loss_clip": 0.01137233,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.04761612,
      "balance_loss_mlp": 1.02003133,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.9845969387751514,
      "language_loss": 0.77753109,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.7991792,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 3.644270420074463
    },
    {
      "auxiliary_loss_clip": 0.01166769,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.05076599,
      "balance_loss_mlp": 1.01703036,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 3.5944272103764274,
      "language_loss": 0.78366506,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80557537,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.6015799045562744
    },
    {
      "auxiliary_loss_clip": 0.01180671,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.05207074,
      "balance_loss_mlp": 1.01569581,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 3.53881314940634,
      "language_loss": 0.83902144,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.86106378,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.5986664295196533
    },
    {
      "auxiliary_loss_clip": 0.01149269,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.04609394,
      "balance_loss_mlp": 1.01965237,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.5151905073393226,
      "language_loss": 0.85036987,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.87213218,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.752962350845337
    },
    {
      "auxiliary_loss_clip": 0.01154352,
      "auxiliary_loss_mlp": 0.01019833,
      "balance_loss_clip": 1.04291153,
      "balance_loss_mlp": 1.01222777,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.7572759965097768,
      "language_loss": 0.83077466,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85251641,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.680331230163574
    },
    {
      "auxiliary_loss_clip": 0.01174037,
      "auxiliary_loss_mlp": 0.01022131,
      "balance_loss_clip": 1.05164766,
      "balance_loss_mlp": 1.01518393,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.7821902947024877,
      "language_loss": 0.79797733,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.81993902,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.5629541873931885
    },
    {
      "auxiliary_loss_clip": 0.01173786,
      "auxiliary_loss_mlp": 0.01022879,
      "balance_loss_clip": 1.0501647,
      "balance_loss_mlp": 1.01527357,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 2.167970847466596,
      "language_loss": 0.67298532,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69495201,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.729058265686035
    },
    {
      "auxiliary_loss_clip": 0.01143069,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.04638624,
      "balance_loss_mlp": 1.0200038,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.1298241697625433,
      "language_loss": 0.78363371,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80534476,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.6958487033843994
    },
    {
      "auxiliary_loss_clip": 0.0116339,
      "auxiliary_loss_mlp": 0.01024458,
      "balance_loss_clip": 1.04921389,
      "balance_loss_mlp": 1.01713288,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.3267914923018123,
      "language_loss": 0.75326383,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.77514231,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.7475368976593018
    },
    {
      "auxiliary_loss_clip": 0.01152427,
      "auxiliary_loss_mlp": 0.01023796,
      "balance_loss_clip": 1.04876065,
      "balance_loss_mlp": 1.0163188,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.5819363305584961,
      "language_loss": 0.70013475,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72189701,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 3.5312695503234863
    },
    {
      "auxiliary_loss_clip": 0.01136166,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04586112,
      "balance_loss_mlp": 1.01984119,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.6708655990127952,
      "language_loss": 0.68368709,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70532542,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.816404342651367
    },
    {
      "auxiliary_loss_clip": 0.01169277,
      "auxiliary_loss_mlp": 0.01023159,
      "balance_loss_clip": 1.05131745,
      "balance_loss_mlp": 1.01618564,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 2.3029816758996353,
      "language_loss": 0.81156409,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.8334884,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.6006836891174316
    },
    {
      "auxiliary_loss_clip": 0.01176885,
      "auxiliary_loss_mlp": 0.01027933,
      "balance_loss_clip": 1.05329335,
      "balance_loss_mlp": 1.02080417,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 3.1068560289281457,
      "language_loss": 0.81354403,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83559227,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.532785177230835
    },
    {
      "auxiliary_loss_clip": 0.01173358,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.04833579,
      "balance_loss_mlp": 1.01658762,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 2.0591229655098693,
      "language_loss": 0.74061406,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76258814,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 3.483372211456299
    },
    {
      "auxiliary_loss_clip": 0.01175727,
      "auxiliary_loss_mlp": 0.01024602,
      "balance_loss_clip": 1.05300713,
      "balance_loss_mlp": 1.01744366,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 2.039737785941035,
      "language_loss": 0.76745081,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78945416,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.5348739624023438
    },
    {
      "auxiliary_loss_clip": 0.01072372,
      "auxiliary_loss_mlp": 0.0100302,
      "balance_loss_clip": 1.02116287,
      "balance_loss_mlp": 1.00211954,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8103984105991535,
      "language_loss": 0.62484503,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64559889,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.260594606399536
    },
    {
      "auxiliary_loss_clip": 0.01144419,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.04474509,
      "balance_loss_mlp": 1.01940298,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.8321050850008709,
      "language_loss": 0.7970531,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81877345,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.7390835285186768
    },
    {
      "auxiliary_loss_clip": 0.01175053,
      "auxiliary_loss_mlp": 0.01021727,
      "balance_loss_clip": 1.05024672,
      "balance_loss_mlp": 1.01506615,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 1.960596625216698,
      "language_loss": 0.72387087,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74583864,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.7268877029418945
    },
    {
      "auxiliary_loss_clip": 0.01105729,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.03993356,
      "balance_loss_mlp": 1.02087605,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.3473451359219393,
      "language_loss": 0.83974606,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.86109334,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.900214195251465
    },
    {
      "auxiliary_loss_clip": 0.01155521,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04960632,
      "balance_loss_mlp": 1.01965499,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 2.591453766901079,
      "language_loss": 0.75581372,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77764142,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 3.824641466140747
    },
    {
      "auxiliary_loss_clip": 0.01127618,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.04502916,
      "balance_loss_mlp": 1.02190709,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.967826061927155,
      "language_loss": 0.74755156,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76911968,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.6689374446868896
    },
    {
      "auxiliary_loss_clip": 0.01174023,
      "auxiliary_loss_mlp": 0.01023893,
      "balance_loss_clip": 1.05063415,
      "balance_loss_mlp": 1.01691937,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 1.7953248913048498,
      "language_loss": 0.78031558,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80229473,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.6279656887054443
    },
    {
      "auxiliary_loss_clip": 0.01083294,
      "auxiliary_loss_mlp": 0.01001129,
      "balance_loss_clip": 1.02012944,
      "balance_loss_mlp": 1.0002116,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.7845374657859377,
      "language_loss": 0.57514203,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59598625,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.183943748474121
    },
    {
      "auxiliary_loss_clip": 0.01152398,
      "auxiliary_loss_mlp": 0.01023328,
      "balance_loss_clip": 1.04568326,
      "balance_loss_mlp": 1.0157938,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 3.572022433250727,
      "language_loss": 0.81440842,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83616573,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.667451858520508
    },
    {
      "auxiliary_loss_clip": 0.01179223,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.05307066,
      "balance_loss_mlp": 1.02318275,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 2.0563558423345056,
      "language_loss": 0.72248167,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74458754,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.5534911155700684
    },
    {
      "auxiliary_loss_clip": 0.01156665,
      "auxiliary_loss_mlp": 0.01022938,
      "balance_loss_clip": 1.0482136,
      "balance_loss_mlp": 1.01599956,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.038863578090858,
      "language_loss": 0.71850944,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74030548,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.6214919090270996
    },
    {
      "auxiliary_loss_clip": 0.01151391,
      "auxiliary_loss_mlp": 0.00901221,
      "balance_loss_clip": 1.04545581,
      "balance_loss_mlp": 1.00077868,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.423396222119267,
      "language_loss": 0.80554682,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.82607299,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.624098062515259
    },
    {
      "auxiliary_loss_clip": 0.01068577,
      "auxiliary_loss_mlp": 0.0100174,
      "balance_loss_clip": 1.019508,
      "balance_loss_mlp": 1.00070882,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7717189573980497,
      "language_loss": 0.59656179,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61726493,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 4.151246786117554
    },
    {
      "auxiliary_loss_clip": 0.01077827,
      "auxiliary_loss_mlp": 0.0100146,
      "balance_loss_clip": 1.01856494,
      "balance_loss_mlp": 1.00057828,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8527982177569049,
      "language_loss": 0.5509938,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57178664,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.1782238483428955
    },
    {
      "auxiliary_loss_clip": 0.01154929,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.04724729,
      "balance_loss_mlp": 1.01761413,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.9023711739049383,
      "language_loss": 0.82454991,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.84634638,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.6102375984191895
    },
    {
      "auxiliary_loss_clip": 0.01165276,
      "auxiliary_loss_mlp": 0.01026875,
      "balance_loss_clip": 1.05239427,
      "balance_loss_mlp": 1.0201335,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 2.0893239949765308,
      "language_loss": 0.84361041,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86553192,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.7089273929595947
    },
    {
      "auxiliary_loss_clip": 0.01148326,
      "auxiliary_loss_mlp": 0.01021356,
      "balance_loss_clip": 1.04605675,
      "balance_loss_mlp": 1.01423621,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 2.407466329218278,
      "language_loss": 0.72254306,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74423993,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.716773748397827
    },
    {
      "auxiliary_loss_clip": 0.01146032,
      "auxiliary_loss_mlp": 0.01021341,
      "balance_loss_clip": 1.04461455,
      "balance_loss_mlp": 1.01427805,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 3.4449074718453767,
      "language_loss": 0.79773456,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81940824,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7332632541656494
    },
    {
      "auxiliary_loss_clip": 0.01159651,
      "auxiliary_loss_mlp": 0.01022812,
      "balance_loss_clip": 1.04800904,
      "balance_loss_mlp": 1.01614785,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.8840732744680648,
      "language_loss": 0.73119426,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75301886,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.6052372455596924
    },
    {
      "auxiliary_loss_clip": 0.01162459,
      "auxiliary_loss_mlp": 0.01023944,
      "balance_loss_clip": 1.04805422,
      "balance_loss_mlp": 1.01631474,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.399606357363176,
      "language_loss": 0.76376092,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78562486,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.6555962562561035
    },
    {
      "auxiliary_loss_clip": 0.01154399,
      "auxiliary_loss_mlp": 0.01026018,
      "balance_loss_clip": 1.0513742,
      "balance_loss_mlp": 1.01854348,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 2.4761784984548862,
      "language_loss": 0.73566449,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.7574687,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.6325247287750244
    },
    {
      "auxiliary_loss_clip": 0.01151445,
      "auxiliary_loss_mlp": 0.01023189,
      "balance_loss_clip": 1.04686606,
      "balance_loss_mlp": 1.01593804,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.9423940409445741,
      "language_loss": 0.55096704,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57271338,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 3.765559434890747
    },
    {
      "auxiliary_loss_clip": 0.01175675,
      "auxiliary_loss_mlp": 0.01024504,
      "balance_loss_clip": 1.05123842,
      "balance_loss_mlp": 1.01707768,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.5977284105886165,
      "language_loss": 0.77507985,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79708159,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.590484380722046
    },
    {
      "auxiliary_loss_clip": 0.01154691,
      "auxiliary_loss_mlp": 0.01023056,
      "balance_loss_clip": 1.04631448,
      "balance_loss_mlp": 1.01597166,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 3.18560131575885,
      "language_loss": 0.76340204,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.7851795,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.6635689735412598
    },
    {
      "auxiliary_loss_clip": 0.01161806,
      "auxiliary_loss_mlp": 0.01023076,
      "balance_loss_clip": 1.04748368,
      "balance_loss_mlp": 1.01503563,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 3.1241052943840852,
      "language_loss": 0.76318359,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78503245,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.72818922996521
    },
    {
      "auxiliary_loss_clip": 0.01157916,
      "auxiliary_loss_mlp": 0.01027448,
      "balance_loss_clip": 1.04920852,
      "balance_loss_mlp": 1.02005684,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.9081212901485713,
      "language_loss": 0.8112092,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83306277,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 3.634552240371704
    },
    {
      "auxiliary_loss_clip": 0.01157686,
      "auxiliary_loss_mlp": 0.01029229,
      "balance_loss_clip": 1.04688835,
      "balance_loss_mlp": 1.02111709,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.437988316486842,
      "language_loss": 0.73748457,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.7593537,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.635115623474121
    },
    {
      "auxiliary_loss_clip": 0.01148221,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.04581285,
      "balance_loss_mlp": 1.01938915,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 3.113141902021554,
      "language_loss": 0.89588141,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.917633,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.7121078968048096
    },
    {
      "auxiliary_loss_clip": 0.01144281,
      "auxiliary_loss_mlp": 0.01021705,
      "balance_loss_clip": 1.04756117,
      "balance_loss_mlp": 1.01497555,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 2.263435332594279,
      "language_loss": 0.87034416,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.89200401,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.7127444744110107
    },
    {
      "auxiliary_loss_clip": 0.01171877,
      "auxiliary_loss_mlp": 0.01025223,
      "balance_loss_clip": 1.05532026,
      "balance_loss_mlp": 1.0170511,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 3.6600970613295636,
      "language_loss": 0.68296862,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.7049396,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.6841259002685547
    },
    {
      "auxiliary_loss_clip": 0.01154352,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04782701,
      "balance_loss_mlp": 1.01729965,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.91553376127586,
      "language_loss": 0.78517592,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.8069728,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.6843509674072266
    },
    {
      "auxiliary_loss_clip": 0.01149222,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 1.04844224,
      "balance_loss_mlp": 1.01960945,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.8349906256299207,
      "language_loss": 0.80159068,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82335716,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 3.9539175033569336
    },
    {
      "auxiliary_loss_clip": 0.0116165,
      "auxiliary_loss_mlp": 0.01021807,
      "balance_loss_clip": 1.04937077,
      "balance_loss_mlp": 1.01394534,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.65800788586734,
      "language_loss": 0.83116698,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85300159,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.7388267517089844
    },
    {
      "auxiliary_loss_clip": 0.01147281,
      "auxiliary_loss_mlp": 0.01019622,
      "balance_loss_clip": 1.04385853,
      "balance_loss_mlp": 1.01160502,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.438992704488684,
      "language_loss": 0.65878212,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.68045115,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.685002326965332
    },
    {
      "auxiliary_loss_clip": 0.0105565,
      "auxiliary_loss_mlp": 0.01008214,
      "balance_loss_clip": 1.01868415,
      "balance_loss_mlp": 1.00723648,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7966333526918351,
      "language_loss": 0.56546932,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58610797,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.201139211654663
    },
    {
      "auxiliary_loss_clip": 0.0118186,
      "auxiliary_loss_mlp": 0.01022433,
      "balance_loss_clip": 1.05516601,
      "balance_loss_mlp": 1.01456499,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 3.1423004189094916,
      "language_loss": 0.8666842,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88872719,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.574413537979126
    },
    {
      "auxiliary_loss_clip": 0.01152815,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.04587126,
      "balance_loss_mlp": 1.01726699,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.7342848378286213,
      "language_loss": 0.64064676,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66242182,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.7903494834899902
    },
    {
      "auxiliary_loss_clip": 0.01164052,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.04978895,
      "balance_loss_mlp": 1.01735044,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.646982176843379,
      "language_loss": 0.70950043,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.73139322,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.752439498901367
    },
    {
      "auxiliary_loss_clip": 0.01160444,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04758871,
      "balance_loss_mlp": 1.01728964,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 1.9341642851824583,
      "language_loss": 0.70191991,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72377402,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.635108709335327
    },
    {
      "auxiliary_loss_clip": 0.01145294,
      "auxiliary_loss_mlp": 0.01026996,
      "balance_loss_clip": 1.04756021,
      "balance_loss_mlp": 1.01984382,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.601854069032306,
      "language_loss": 0.75732625,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.77904916,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 3.728567123413086
    },
    {
      "auxiliary_loss_clip": 0.01158921,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.05175674,
      "balance_loss_mlp": 1.01881623,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 2.0640224258109026,
      "language_loss": 0.76096225,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.78281057,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.6590585708618164
    },
    {
      "auxiliary_loss_clip": 0.01174858,
      "auxiliary_loss_mlp": 0.01026206,
      "balance_loss_clip": 1.05120015,
      "balance_loss_mlp": 1.0193398,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.8117226649544893,
      "language_loss": 0.71216422,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73417485,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.694099187850952
    },
    {
      "auxiliary_loss_clip": 0.01168609,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.04769897,
      "balance_loss_mlp": 1.01855969,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 1.5819398044906903,
      "language_loss": 0.78862286,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.81056643,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 2.6254947185516357
    },
    {
      "auxiliary_loss_clip": 0.01137394,
      "auxiliary_loss_mlp": 0.01022275,
      "balance_loss_clip": 1.0452503,
      "balance_loss_mlp": 1.01513112,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 2.051133542034038,
      "language_loss": 0.77902693,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80062366,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.718240976333618
    },
    {
      "auxiliary_loss_clip": 0.01152161,
      "auxiliary_loss_mlp": 0.01024335,
      "balance_loss_clip": 1.04545379,
      "balance_loss_mlp": 1.01731944,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 2.2895002095324744,
      "language_loss": 0.73977649,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.76154149,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.668531656265259
    },
    {
      "auxiliary_loss_clip": 0.01111609,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.04325557,
      "balance_loss_mlp": 1.01970816,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.2366192566291283,
      "language_loss": 0.62587833,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64726228,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 3.115657329559326
    },
    {
      "auxiliary_loss_clip": 0.01149874,
      "auxiliary_loss_mlp": 0.01023067,
      "balance_loss_clip": 1.04592478,
      "balance_loss_mlp": 1.01577771,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.2613580328302287,
      "language_loss": 0.73885095,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76058036,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 3.1506054401397705
    },
    {
      "auxiliary_loss_clip": 0.01159214,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.04897213,
      "balance_loss_mlp": 1.01858854,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.69791410569432,
      "language_loss": 0.80819273,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.8300463,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 3.6921513080596924
    },
    {
      "auxiliary_loss_clip": 0.01162841,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.04636407,
      "balance_loss_mlp": 1.01652825,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.9759920097271748,
      "language_loss": 0.8313207,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85319114,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.6462152004241943
    },
    {
      "auxiliary_loss_clip": 0.0116371,
      "auxiliary_loss_mlp": 0.01024151,
      "balance_loss_clip": 1.05111897,
      "balance_loss_mlp": 1.01715064,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 2.103124206079643,
      "language_loss": 0.63520193,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.65708053,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.6515908241271973
    },
    {
      "auxiliary_loss_clip": 0.01083472,
      "auxiliary_loss_mlp": 0.01001069,
      "balance_loss_clip": 1.01988411,
      "balance_loss_mlp": 1.00015724,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7829585834963542,
      "language_loss": 0.55933052,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58017594,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.0510077476501465
    },
    {
      "auxiliary_loss_clip": 0.01176864,
      "auxiliary_loss_mlp": 0.01021524,
      "balance_loss_clip": 1.05155694,
      "balance_loss_mlp": 1.01377261,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 1.9937851453881728,
      "language_loss": 0.71196204,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73394597,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 3.485347270965576
    },
    {
      "auxiliary_loss_clip": 0.01173368,
      "auxiliary_loss_mlp": 0.01023188,
      "balance_loss_clip": 1.05173683,
      "balance_loss_mlp": 1.01516497,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 2.183249012957699,
      "language_loss": 0.7594139,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.78137946,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.641507863998413
    },
    {
      "auxiliary_loss_clip": 0.01162954,
      "auxiliary_loss_mlp": 0.01022696,
      "balance_loss_clip": 1.05203533,
      "balance_loss_mlp": 1.01515055,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.7100601624203795,
      "language_loss": 0.7239877,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.74584413,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.7486038208007812
    },
    {
      "auxiliary_loss_clip": 0.01162077,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.05047441,
      "balance_loss_mlp": 1.01986969,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 3.943975036905753,
      "language_loss": 0.78992516,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.81181759,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.5725884437561035
    },
    {
      "auxiliary_loss_clip": 0.01169548,
      "auxiliary_loss_mlp": 0.01019758,
      "balance_loss_clip": 1.04818368,
      "balance_loss_mlp": 1.01213169,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.198051010083503,
      "language_loss": 0.63950503,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.66139805,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.6741740703582764
    },
    {
      "auxiliary_loss_clip": 0.01165257,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05541992,
      "balance_loss_mlp": 1.01910996,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 2.088142016486164,
      "language_loss": 0.69119078,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71310866,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 3.6640050411224365
    },
    {
      "auxiliary_loss_clip": 0.01167046,
      "auxiliary_loss_mlp": 0.0102555,
      "balance_loss_clip": 1.05203891,
      "balance_loss_mlp": 1.01814127,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.795676732624397,
      "language_loss": 0.72503567,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74696165,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.611320972442627
    },
    {
      "auxiliary_loss_clip": 0.01172743,
      "auxiliary_loss_mlp": 0.01023146,
      "balance_loss_clip": 1.05075002,
      "balance_loss_mlp": 1.01619029,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 2.37426457727117,
      "language_loss": 0.63672298,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.65868187,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.6200008392333984
    },
    {
      "auxiliary_loss_clip": 0.01150128,
      "auxiliary_loss_mlp": 0.0102633,
      "balance_loss_clip": 1.04728079,
      "balance_loss_mlp": 1.01862657,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.0367500776391987,
      "language_loss": 0.75474417,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77650875,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.790343761444092
    },
    {
      "auxiliary_loss_clip": 0.01162507,
      "auxiliary_loss_mlp": 0.01024032,
      "balance_loss_clip": 1.05306065,
      "balance_loss_mlp": 1.01718378,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.5203220112250315,
      "language_loss": 0.67409062,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69595599,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.6633341312408447
    },
    {
      "auxiliary_loss_clip": 0.0114787,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04549325,
      "balance_loss_mlp": 1.01673532,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.7468473094472756,
      "language_loss": 0.7349062,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75662744,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.6057560443878174
    },
    {
      "auxiliary_loss_clip": 0.01137004,
      "auxiliary_loss_mlp": 0.0102236,
      "balance_loss_clip": 1.04285336,
      "balance_loss_mlp": 1.01524878,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.6469548850546747,
      "language_loss": 0.74388433,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76547796,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.7370846271514893
    },
    {
      "auxiliary_loss_clip": 0.01165548,
      "auxiliary_loss_mlp": 0.01024259,
      "balance_loss_clip": 1.05003858,
      "balance_loss_mlp": 1.01708567,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.2633510798377277,
      "language_loss": 0.78136301,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.8032611,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.564096212387085
    },
    {
      "auxiliary_loss_clip": 0.01155654,
      "auxiliary_loss_mlp": 0.01024913,
      "balance_loss_clip": 1.05130625,
      "balance_loss_mlp": 1.01693833,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.5238730921392145,
      "language_loss": 0.72517371,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74697936,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.6582794189453125
    },
    {
      "auxiliary_loss_clip": 0.01154348,
      "auxiliary_loss_mlp": 0.01035177,
      "balance_loss_clip": 1.05363488,
      "balance_loss_mlp": 1.02710092,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 2.1785107066325526,
      "language_loss": 0.71529341,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73718858,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 4.067829608917236
    },
    {
      "auxiliary_loss_clip": 0.01174012,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.04772079,
      "balance_loss_mlp": 1.0184015,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.4798958448193194,
      "language_loss": 0.77405179,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79604656,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.5440025329589844
    },
    {
      "auxiliary_loss_clip": 0.01177111,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.05202425,
      "balance_loss_mlp": 1.02012706,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.4163275506213666,
      "language_loss": 0.81996858,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84202182,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.736424684524536
    },
    {
      "auxiliary_loss_clip": 0.01168549,
      "auxiliary_loss_mlp": 0.01022526,
      "balance_loss_clip": 1.05133605,
      "balance_loss_mlp": 1.01522088,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.123959814461896,
      "language_loss": 0.7410531,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76296389,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.6486306190490723
    },
    {
      "auxiliary_loss_clip": 0.01176124,
      "auxiliary_loss_mlp": 0.00901177,
      "balance_loss_clip": 1.05066705,
      "balance_loss_mlp": 1.0008322,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 3.083100170631487,
      "language_loss": 0.84350157,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86427462,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.5561985969543457
    },
    {
      "auxiliary_loss_clip": 0.01168806,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.05003846,
      "balance_loss_mlp": 1.01991117,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.8049748000614474,
      "language_loss": 0.77664804,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79860413,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.682765007019043
    },
    {
      "auxiliary_loss_clip": 0.01140265,
      "auxiliary_loss_mlp": 0.01027638,
      "balance_loss_clip": 1.04489028,
      "balance_loss_mlp": 1.0202415,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.7231358421659504,
      "language_loss": 0.75538313,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77706218,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.809206962585449
    },
    {
      "auxiliary_loss_clip": 0.01165544,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 1.04818082,
      "balance_loss_mlp": 1.02584028,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 3.0964696295610206,
      "language_loss": 0.78742778,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80940998,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.6153182983398438
    },
    {
      "auxiliary_loss_clip": 0.01163018,
      "auxiliary_loss_mlp": 0.01023632,
      "balance_loss_clip": 1.04896498,
      "balance_loss_mlp": 1.01644397,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.7087250286830697,
      "language_loss": 0.88530779,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90717423,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.6649396419525146
    },
    {
      "auxiliary_loss_clip": 0.01161815,
      "auxiliary_loss_mlp": 0.01026622,
      "balance_loss_clip": 1.04921234,
      "balance_loss_mlp": 1.01913261,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 1.720787885566224,
      "language_loss": 0.69985861,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72174299,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 3.5388941764831543
    },
    {
      "auxiliary_loss_clip": 0.01168425,
      "auxiliary_loss_mlp": 0.01025683,
      "balance_loss_clip": 1.05048537,
      "balance_loss_mlp": 1.01801205,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.059002484927634,
      "language_loss": 0.74439418,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76633531,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.561897039413452
    },
    {
      "auxiliary_loss_clip": 0.01166032,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.04801619,
      "balance_loss_mlp": 1.01857781,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.5667728084917187,
      "language_loss": 0.80539542,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82731783,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.5941860675811768
    },
    {
      "auxiliary_loss_clip": 0.01146704,
      "auxiliary_loss_mlp": 0.01019525,
      "balance_loss_clip": 1.04704905,
      "balance_loss_mlp": 1.01257467,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 2.233283999199245,
      "language_loss": 0.78674579,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.80840808,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.694251298904419
    },
    {
      "auxiliary_loss_clip": 0.01127017,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.04476142,
      "balance_loss_mlp": 1.01666534,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.8706340440581053,
      "language_loss": 0.70878184,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.7302916,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 3.860232353210449
    },
    {
      "auxiliary_loss_clip": 0.01161997,
      "auxiliary_loss_mlp": 0.00901181,
      "balance_loss_clip": 1.04887331,
      "balance_loss_mlp": 1.00081825,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 1.6926276219823195,
      "language_loss": 0.73363495,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75426674,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.7940516471862793
    },
    {
      "auxiliary_loss_clip": 0.01174247,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.05039167,
      "balance_loss_mlp": 1.0178647,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.8952790666837955,
      "language_loss": 0.70345581,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72545224,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.603422164916992
    },
    {
      "auxiliary_loss_clip": 0.01169764,
      "auxiliary_loss_mlp": 0.01024879,
      "balance_loss_clip": 1.051103,
      "balance_loss_mlp": 1.01735711,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.456363509143221,
      "language_loss": 0.65542966,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67737615,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.636119842529297
    },
    {
      "auxiliary_loss_clip": 0.01157191,
      "auxiliary_loss_mlp": 0.01023239,
      "balance_loss_clip": 1.04816675,
      "balance_loss_mlp": 1.01569915,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 1.8986351589982704,
      "language_loss": 0.74355966,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76536405,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.6751723289489746
    },
    {
      "auxiliary_loss_clip": 0.01154524,
      "auxiliary_loss_mlp": 0.00901287,
      "balance_loss_clip": 1.05262256,
      "balance_loss_mlp": 1.00065351,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 2.133227828235101,
      "language_loss": 0.73287755,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75343573,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 3.5909924507141113
    },
    {
      "auxiliary_loss_clip": 0.01158646,
      "auxiliary_loss_mlp": 0.01027971,
      "balance_loss_clip": 1.05007243,
      "balance_loss_mlp": 1.0201093,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 2.038317835509807,
      "language_loss": 0.80579805,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82766414,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.6131017208099365
    },
    {
      "auxiliary_loss_clip": 0.0115858,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.04433155,
      "balance_loss_mlp": 1.01583099,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.5542505889894667,
      "language_loss": 0.77061558,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79243392,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.699414014816284
    },
    {
      "auxiliary_loss_clip": 0.01116537,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.048769,
      "balance_loss_mlp": 1.0165143,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 2.2421191572954906,
      "language_loss": 0.6617341,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68314576,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.779977798461914
    },
    {
      "auxiliary_loss_clip": 0.01165274,
      "auxiliary_loss_mlp": 0.01020353,
      "balance_loss_clip": 1.04961014,
      "balance_loss_mlp": 1.01326561,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.252179108443893,
      "language_loss": 0.67754412,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69940042,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.690810441970825
    },
    {
      "auxiliary_loss_clip": 0.01151625,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.04705787,
      "balance_loss_mlp": 1.02067363,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 1.9930295671455478,
      "language_loss": 0.7884661,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81026661,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.736283540725708
    },
    {
      "auxiliary_loss_clip": 0.01154028,
      "auxiliary_loss_mlp": 0.01028732,
      "balance_loss_clip": 1.04979372,
      "balance_loss_mlp": 1.0210247,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 4.981441882566863,
      "language_loss": 0.68158549,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70341313,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.7385849952697754
    },
    {
      "auxiliary_loss_clip": 0.01056148,
      "auxiliary_loss_mlp": 0.00891387,
      "balance_loss_clip": 1.01808262,
      "balance_loss_mlp": 1.00004661,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7476297607066675,
      "language_loss": 0.55777299,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57724833,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.197047233581543
    },
    {
      "auxiliary_loss_clip": 0.01162475,
      "auxiliary_loss_mlp": 0.01023253,
      "balance_loss_clip": 1.05037928,
      "balance_loss_mlp": 1.01590109,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.6674322607715062,
      "language_loss": 0.76943827,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79129559,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.6594295501708984
    },
    {
      "auxiliary_loss_clip": 0.01174628,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.05104804,
      "balance_loss_mlp": 1.01765704,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.2487772385332296,
      "language_loss": 0.77474487,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.7967419,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 3.833203077316284
    },
    {
      "auxiliary_loss_clip": 0.01156245,
      "auxiliary_loss_mlp": 0.01022646,
      "balance_loss_clip": 1.05115426,
      "balance_loss_mlp": 1.01482272,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.2563203011887016,
      "language_loss": 0.82725561,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.8490445,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.7007863521575928
    },
    {
      "auxiliary_loss_clip": 0.01107026,
      "auxiliary_loss_mlp": 0.01019802,
      "balance_loss_clip": 1.03859472,
      "balance_loss_mlp": 1.01231599,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.136490394447303,
      "language_loss": 0.74384582,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.76511407,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.825521945953369
    },
    {
      "auxiliary_loss_clip": 0.01122985,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.03849292,
      "balance_loss_mlp": 1.01895463,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 2.867455346075327,
      "language_loss": 0.7863239,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80781734,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.7876107692718506
    },
    {
      "auxiliary_loss_clip": 0.01153501,
      "auxiliary_loss_mlp": 0.01022962,
      "balance_loss_clip": 1.04496729,
      "balance_loss_mlp": 1.01431394,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.0520668416847454,
      "language_loss": 0.71718168,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.73894632,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7336652278900146
    },
    {
      "auxiliary_loss_clip": 0.01167062,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.05243671,
      "balance_loss_mlp": 1.0186491,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.09830147642787,
      "language_loss": 0.73869944,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.76062638,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.7079384326934814
    },
    {
      "auxiliary_loss_clip": 0.01153268,
      "auxiliary_loss_mlp": 0.01020439,
      "balance_loss_clip": 1.04581642,
      "balance_loss_mlp": 1.01269078,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.2115005922014634,
      "language_loss": 0.63275254,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65448964,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.6742217540740967
    },
    {
      "auxiliary_loss_clip": 0.0113247,
      "auxiliary_loss_mlp": 0.00901516,
      "balance_loss_clip": 1.04184604,
      "balance_loss_mlp": 1.00073874,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 3.102750719656751,
      "language_loss": 0.75463784,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77497768,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.7115213871002197
    },
    {
      "auxiliary_loss_clip": 0.01157739,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.04436326,
      "balance_loss_mlp": 1.02117205,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.613931947641981,
      "language_loss": 0.78473985,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80661178,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.733213186264038
    },
    {
      "auxiliary_loss_clip": 0.0116276,
      "auxiliary_loss_mlp": 0.01029852,
      "balance_loss_clip": 1.04880881,
      "balance_loss_mlp": 1.02232361,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 2.1407097612639094,
      "language_loss": 0.71702933,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.7389555,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 3.5199522972106934
    },
    {
      "auxiliary_loss_clip": 0.01145909,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 1.04464912,
      "balance_loss_mlp": 1.01944089,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.182729469982647,
      "language_loss": 0.62540185,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.6471343,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.713979721069336
    },
    {
      "auxiliary_loss_clip": 0.01161364,
      "auxiliary_loss_mlp": 0.01030295,
      "balance_loss_clip": 1.05051041,
      "balance_loss_mlp": 1.0227077,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.2953459662000504,
      "language_loss": 0.798388,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.82030463,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.626005172729492
    },
    {
      "auxiliary_loss_clip": 0.01155443,
      "auxiliary_loss_mlp": 0.01027054,
      "balance_loss_clip": 1.04957986,
      "balance_loss_mlp": 1.02033091,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 2.173441617945145,
      "language_loss": 0.71860415,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.7404291,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 3.607473134994507
    },
    {
      "auxiliary_loss_clip": 0.0108286,
      "auxiliary_loss_mlp": 0.01003358,
      "balance_loss_clip": 1.01934648,
      "balance_loss_mlp": 1.00245786,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.802263657247363,
      "language_loss": 0.62365508,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.6445173,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.2095658779144287
    },
    {
      "auxiliary_loss_clip": 0.0116576,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.05245805,
      "balance_loss_mlp": 1.01880836,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 1.8332596613559786,
      "language_loss": 0.66834134,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.6902616,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.585296154022217
    },
    {
      "auxiliary_loss_clip": 0.01181621,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 1.05291247,
      "balance_loss_mlp": 1.01988554,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.9205692925702285,
      "language_loss": 0.79917288,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.82126695,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.5787434577941895
    },
    {
      "auxiliary_loss_clip": 0.01170368,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.05005026,
      "balance_loss_mlp": 1.01752973,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 2.0465920969430718,
      "language_loss": 0.72680175,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.74875653,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.6107261180877686
    },
    {
      "auxiliary_loss_clip": 0.01143204,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.04476488,
      "balance_loss_mlp": 1.02072537,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 2.4498374756242627,
      "language_loss": 0.72317481,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74489492,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.7190117835998535
    },
    {
      "auxiliary_loss_clip": 0.01137117,
      "auxiliary_loss_mlp": 0.01023433,
      "balance_loss_clip": 1.04497623,
      "balance_loss_mlp": 1.01592875,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.688366242944186,
      "language_loss": 0.76306736,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78467292,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 3.6225123405456543
    },
    {
      "auxiliary_loss_clip": 0.01136062,
      "auxiliary_loss_mlp": 0.01027589,
      "balance_loss_clip": 1.04106903,
      "balance_loss_mlp": 1.0196557,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 2.0791881495577993,
      "language_loss": 0.81920713,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84084362,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.7661073207855225
    },
    {
      "auxiliary_loss_clip": 0.01177712,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.05075514,
      "balance_loss_mlp": 1.02257013,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.618024205879185,
      "language_loss": 0.84149319,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86357772,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.5231006145477295
    },
    {
      "auxiliary_loss_clip": 0.01142417,
      "auxiliary_loss_mlp": 0.01021533,
      "balance_loss_clip": 1.04497552,
      "balance_loss_mlp": 1.01390028,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.7701078443117315,
      "language_loss": 0.75658906,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77822852,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.7326440811157227
    },
    {
      "auxiliary_loss_clip": 0.01143628,
      "auxiliary_loss_mlp": 0.01023506,
      "balance_loss_clip": 1.04680777,
      "balance_loss_mlp": 1.01559007,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.4231472288915104,
      "language_loss": 0.74615854,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.7678299,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.717604160308838
    },
    {
      "auxiliary_loss_clip": 0.01178237,
      "auxiliary_loss_mlp": 0.0102487,
      "balance_loss_clip": 1.05401254,
      "balance_loss_mlp": 1.01691937,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.0305630845543035,
      "language_loss": 0.82969075,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.85172182,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.636892318725586
    },
    {
      "auxiliary_loss_clip": 0.01169284,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.04837644,
      "balance_loss_mlp": 1.01861393,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.4172070155004373,
      "language_loss": 0.77884841,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.8008076,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.6177010536193848
    },
    {
      "auxiliary_loss_clip": 0.01153189,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.04881835,
      "balance_loss_mlp": 1.01905668,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.8483465070395928,
      "language_loss": 0.76646829,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.7882607,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.702810764312744
    },
    {
      "auxiliary_loss_clip": 0.01162167,
      "auxiliary_loss_mlp": 0.01026461,
      "balance_loss_clip": 1.04895663,
      "balance_loss_mlp": 1.01901674,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 2.7819171711030606,
      "language_loss": 0.80955285,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.83143914,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 3.546783685684204
    },
    {
      "auxiliary_loss_clip": 0.01164282,
      "auxiliary_loss_mlp": 0.01025977,
      "balance_loss_clip": 1.05060506,
      "balance_loss_mlp": 1.01828182,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 3.7039109756048334,
      "language_loss": 0.64000344,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66190606,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.8209426403045654
    },
    {
      "auxiliary_loss_clip": 0.01159135,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.04868174,
      "balance_loss_mlp": 1.02147317,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.663590096235996,
      "language_loss": 0.6895839,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71146131,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.67307186126709
    },
    {
      "auxiliary_loss_clip": 0.01111437,
      "auxiliary_loss_mlp": 0.00901354,
      "balance_loss_clip": 1.04150915,
      "balance_loss_mlp": 1.00077534,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 1.9293576595481456,
      "language_loss": 0.78582537,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80595332,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 3.0311667919158936
    },
    {
      "auxiliary_loss_clip": 0.01143538,
      "auxiliary_loss_mlp": 0.01032006,
      "balance_loss_clip": 1.04630423,
      "balance_loss_mlp": 1.02388799,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.620204020250668,
      "language_loss": 0.81067884,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83243424,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 2.979356288909912
    },
    {
      "auxiliary_loss_clip": 0.01173383,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.05130339,
      "balance_loss_mlp": 1.0182606,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 2.05567445559872,
      "language_loss": 0.75011939,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.77210903,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.762920379638672
    },
    {
      "auxiliary_loss_clip": 0.01155726,
      "auxiliary_loss_mlp": 0.01024959,
      "balance_loss_clip": 1.04697394,
      "balance_loss_mlp": 1.01698983,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 8.963698124006013,
      "language_loss": 0.69234133,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71414816,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.680241346359253
    },
    {
      "auxiliary_loss_clip": 0.01137745,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.04167271,
      "balance_loss_mlp": 1.01691008,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.673006385905055,
      "language_loss": 0.72253776,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74415976,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.6958913803100586
    },
    {
      "auxiliary_loss_clip": 0.01126278,
      "auxiliary_loss_mlp": 0.01019841,
      "balance_loss_clip": 1.04440737,
      "balance_loss_mlp": 1.01281381,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 2.162593256003028,
      "language_loss": 0.78878319,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.81024438,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.670058012008667
    },
    {
      "auxiliary_loss_clip": 0.01173231,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.05108213,
      "balance_loss_mlp": 1.02042902,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.8019865240192765,
      "language_loss": 0.74068606,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76269358,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 3.4955437183380127
    },
    {
      "auxiliary_loss_clip": 0.01139923,
      "auxiliary_loss_mlp": 0.00900542,
      "balance_loss_clip": 1.04670835,
      "balance_loss_mlp": 1.00078535,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.6549604192580798,
      "language_loss": 0.80306554,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.82347023,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.7764573097229004
    },
    {
      "auxiliary_loss_clip": 0.01165336,
      "auxiliary_loss_mlp": 0.01027609,
      "balance_loss_clip": 1.04983234,
      "balance_loss_mlp": 1.02014673,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.8638283679863845,
      "language_loss": 0.75619066,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77812016,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.7237210273742676
    },
    {
      "auxiliary_loss_clip": 0.01157403,
      "auxiliary_loss_mlp": 0.0103014,
      "balance_loss_clip": 1.05037272,
      "balance_loss_mlp": 1.02303505,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.221415393436054,
      "language_loss": 0.71880507,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.74068052,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.674208164215088
    },
    {
      "auxiliary_loss_clip": 0.01161063,
      "auxiliary_loss_mlp": 0.01024476,
      "balance_loss_clip": 1.04790998,
      "balance_loss_mlp": 1.01726103,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.969571461979584,
      "language_loss": 0.68969321,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71154857,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 3.6441025733947754
    },
    {
      "auxiliary_loss_clip": 0.0115955,
      "auxiliary_loss_mlp": 0.01021764,
      "balance_loss_clip": 1.0486908,
      "balance_loss_mlp": 1.01431012,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 2.360056638777235,
      "language_loss": 0.79766339,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.81947649,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.6595652103424072
    },
    {
      "auxiliary_loss_clip": 0.01169378,
      "auxiliary_loss_mlp": 0.01026095,
      "balance_loss_clip": 1.0481894,
      "balance_loss_mlp": 1.01775682,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 1.9356786844552638,
      "language_loss": 0.7828064,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80476117,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.6201159954071045
    },
    {
      "auxiliary_loss_clip": 0.01054356,
      "auxiliary_loss_mlp": 0.01002588,
      "balance_loss_clip": 1.01984286,
      "balance_loss_mlp": 1.00155723,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7741406862808432,
      "language_loss": 0.64237475,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.6629442,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.314636468887329
    },
    {
      "auxiliary_loss_clip": 0.01155078,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.04956186,
      "balance_loss_mlp": 1.02081394,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.258040462180122,
      "language_loss": 0.72901297,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75084555,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.6249797344207764
    },
    {
      "auxiliary_loss_clip": 0.01159926,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.0489769,
      "balance_loss_mlp": 1.02167606,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 2.0172258342430234,
      "language_loss": 0.63785785,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65974736,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.629335403442383
    },
    {
      "auxiliary_loss_clip": 0.01159759,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.04452062,
      "balance_loss_mlp": 1.01687074,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.4537427265756775,
      "language_loss": 0.71923155,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74107563,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 3.902953863143921
    },
    {
      "auxiliary_loss_clip": 0.01136805,
      "auxiliary_loss_mlp": 0.01030839,
      "balance_loss_clip": 1.04605615,
      "balance_loss_mlp": 1.0226016,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 1.8684575261829894,
      "language_loss": 0.74900919,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77068567,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.673943519592285
    },
    {
      "auxiliary_loss_clip": 0.01175715,
      "auxiliary_loss_mlp": 0.01025603,
      "balance_loss_clip": 1.05214834,
      "balance_loss_mlp": 1.01812291,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 3.0042891107543404,
      "language_loss": 0.61189604,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63390923,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.561490535736084
    },
    {
      "auxiliary_loss_clip": 0.01153165,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 1.05341887,
      "balance_loss_mlp": 1.01918721,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.806424647774844,
      "language_loss": 0.71596324,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.73776799,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.7079782485961914
    },
    {
      "auxiliary_loss_clip": 0.01151708,
      "auxiliary_loss_mlp": 0.01025156,
      "balance_loss_clip": 1.0473063,
      "balance_loss_mlp": 1.01742554,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.9974548439743813,
      "language_loss": 0.71682394,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73859251,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.7906172275543213
    },
    {
      "auxiliary_loss_clip": 0.01159871,
      "auxiliary_loss_mlp": 0.010261,
      "balance_loss_clip": 1.04697442,
      "balance_loss_mlp": 1.01850009,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.7232509843274895,
      "language_loss": 0.80379969,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82565939,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.7089476585388184
    },
    {
      "auxiliary_loss_clip": 0.01152408,
      "auxiliary_loss_mlp": 0.01027007,
      "balance_loss_clip": 1.04854643,
      "balance_loss_mlp": 1.01955044,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.1385582242281473,
      "language_loss": 0.81939352,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.84118766,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.650423526763916
    },
    {
      "auxiliary_loss_clip": 0.0115612,
      "auxiliary_loss_mlp": 0.01026123,
      "balance_loss_clip": 1.04749119,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 2.1152951579585295,
      "language_loss": 0.67811567,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.69993812,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.7151763439178467
    },
    {
      "auxiliary_loss_clip": 0.01161283,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.04761004,
      "balance_loss_mlp": 1.02254343,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 3.204982614470406,
      "language_loss": 0.79811347,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.8200258,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 3.571338415145874
    },
    {
      "auxiliary_loss_clip": 0.01165614,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.05337167,
      "balance_loss_mlp": 1.0197109,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 1.665824423968746,
      "language_loss": 0.74914527,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.77106935,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.6014676094055176
    },
    {
      "auxiliary_loss_clip": 0.01069073,
      "auxiliary_loss_mlp": 0.01002506,
      "balance_loss_clip": 1.01696241,
      "balance_loss_mlp": 1.00152838,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8954618282426408,
      "language_loss": 0.60884649,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62956232,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.072281837463379
    },
    {
      "auxiliary_loss_clip": 0.0116754,
      "auxiliary_loss_mlp": 0.01030902,
      "balance_loss_clip": 1.05189061,
      "balance_loss_mlp": 1.02242684,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.4626883040363206,
      "language_loss": 0.84258252,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.86456698,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.6015279293060303
    },
    {
      "auxiliary_loss_clip": 0.01146086,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.05189633,
      "balance_loss_mlp": 1.02085423,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.6063349933421347,
      "language_loss": 0.77702641,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.7987771,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.7665257453918457
    },
    {
      "auxiliary_loss_clip": 0.01175314,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.04994535,
      "balance_loss_mlp": 1.01963556,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.182346511500021,
      "language_loss": 0.72979045,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75181496,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.724381446838379
    },
    {
      "auxiliary_loss_clip": 0.01146942,
      "auxiliary_loss_mlp": 0.0103087,
      "balance_loss_clip": 1.04562604,
      "balance_loss_mlp": 1.02323508,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.0379775842394867,
      "language_loss": 0.71230215,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73408026,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.7624995708465576
    },
    {
      "auxiliary_loss_clip": 0.01133591,
      "auxiliary_loss_mlp": 0.01031578,
      "balance_loss_clip": 1.0511508,
      "balance_loss_mlp": 1.0233469,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 2.28610581834759,
      "language_loss": 0.82907403,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.85072577,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.731971502304077
    },
    {
      "auxiliary_loss_clip": 0.01165144,
      "auxiliary_loss_mlp": 0.00901656,
      "balance_loss_clip": 1.0492543,
      "balance_loss_mlp": 1.00083888,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 2.6856333916964714,
      "language_loss": 0.66746843,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68813646,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.6179890632629395
    },
    {
      "auxiliary_loss_clip": 0.01152392,
      "auxiliary_loss_mlp": 0.01021763,
      "balance_loss_clip": 1.04546452,
      "balance_loss_mlp": 1.01377559,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.6246847389184882,
      "language_loss": 0.75414908,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77589059,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 3.6638052463531494
    },
    {
      "auxiliary_loss_clip": 0.01130188,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.04535937,
      "balance_loss_mlp": 1.01927316,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 1.8827580307700515,
      "language_loss": 0.71888852,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.74046433,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.7950143814086914
    },
    {
      "auxiliary_loss_clip": 0.01160579,
      "auxiliary_loss_mlp": 0.0090096,
      "balance_loss_clip": 1.05089307,
      "balance_loss_mlp": 1.00073254,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 2.0087810745313157,
      "language_loss": 0.73104143,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.75165677,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.7819297313690186
    },
    {
      "auxiliary_loss_clip": 0.01158277,
      "auxiliary_loss_mlp": 0.00900736,
      "balance_loss_clip": 1.04741836,
      "balance_loss_mlp": 1.00068784,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.7054011603951886,
      "language_loss": 0.86153638,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.88212657,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.6550846099853516
    },
    {
      "auxiliary_loss_clip": 0.0117582,
      "auxiliary_loss_mlp": 0.01022726,
      "balance_loss_clip": 1.05192697,
      "balance_loss_mlp": 1.0152396,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.624744959464536,
      "language_loss": 0.80714297,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82912844,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 3.5783305168151855
    },
    {
      "auxiliary_loss_clip": 0.01152442,
      "auxiliary_loss_mlp": 0.01025835,
      "balance_loss_clip": 1.04680097,
      "balance_loss_mlp": 1.01885509,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.6638385901770738,
      "language_loss": 0.77912539,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.80090821,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.682765007019043
    },
    {
      "auxiliary_loss_clip": 0.01150009,
      "auxiliary_loss_mlp": 0.01029558,
      "balance_loss_clip": 1.04445422,
      "balance_loss_mlp": 1.02142787,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.650312085103425,
      "language_loss": 0.67518628,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69698191,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.659998655319214
    },
    {
      "auxiliary_loss_clip": 0.01151355,
      "auxiliary_loss_mlp": 0.01026634,
      "balance_loss_clip": 1.05010867,
      "balance_loss_mlp": 1.01945806,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 1.818954598293545,
      "language_loss": 0.7608614,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78264129,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.624809741973877
    },
    {
      "auxiliary_loss_clip": 0.01138766,
      "auxiliary_loss_mlp": 0.01026608,
      "balance_loss_clip": 1.04259956,
      "balance_loss_mlp": 1.01929474,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 3.0725685242080165,
      "language_loss": 0.78948593,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.8111397,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.71441650390625
    },
    {
      "auxiliary_loss_clip": 0.01150493,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.04688764,
      "balance_loss_mlp": 1.01748896,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 2.5286772767405696,
      "language_loss": 0.70700836,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72876382,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 3.678492546081543
    },
    {
      "auxiliary_loss_clip": 0.0114698,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.04463696,
      "balance_loss_mlp": 1.02087331,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.191779069575479,
      "language_loss": 0.78919435,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81094605,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.6342246532440186
    },
    {
      "auxiliary_loss_clip": 0.01157655,
      "auxiliary_loss_mlp": 0.00900706,
      "balance_loss_clip": 1.05043089,
      "balance_loss_mlp": 1.00077653,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 4.310565694986733,
      "language_loss": 0.85799849,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.87858212,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.701460838317871
    },
    {
      "auxiliary_loss_clip": 0.01166649,
      "auxiliary_loss_mlp": 0.01024396,
      "balance_loss_clip": 1.0523237,
      "balance_loss_mlp": 1.01707625,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.2835227079264797,
      "language_loss": 0.74352139,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76543188,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.622720241546631
    },
    {
      "auxiliary_loss_clip": 0.01162575,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.05230713,
      "balance_loss_mlp": 1.02009106,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.1453887405132823,
      "language_loss": 0.77424395,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79614705,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.685199499130249
    },
    {
      "auxiliary_loss_clip": 0.01172318,
      "auxiliary_loss_mlp": 0.01024474,
      "balance_loss_clip": 1.04972887,
      "balance_loss_mlp": 1.01760149,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 2.164038953569912,
      "language_loss": 0.73005605,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.75202394,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.594506025314331
    },
    {
      "auxiliary_loss_clip": 0.01175268,
      "auxiliary_loss_mlp": 0.01028146,
      "balance_loss_clip": 1.05034566,
      "balance_loss_mlp": 1.0207969,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.7627593076092551,
      "language_loss": 0.73479152,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.75682569,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.5993361473083496
    },
    {
      "auxiliary_loss_clip": 0.01164097,
      "auxiliary_loss_mlp": 0.0102419,
      "balance_loss_clip": 1.05163968,
      "balance_loss_mlp": 1.01650071,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.031920167516308,
      "language_loss": 0.8477031,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.86958599,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.6369221210479736
    },
    {
      "auxiliary_loss_clip": 0.01167648,
      "auxiliary_loss_mlp": 0.01023559,
      "balance_loss_clip": 1.04766607,
      "balance_loss_mlp": 1.01613855,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.761322705474216,
      "language_loss": 0.86713839,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88905048,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 3.63638973236084
    },
    {
      "auxiliary_loss_clip": 0.0115974,
      "auxiliary_loss_mlp": 0.01028769,
      "balance_loss_clip": 1.04632795,
      "balance_loss_mlp": 1.02176309,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 2.2012529159973075,
      "language_loss": 0.59967685,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62156194,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.6706666946411133
    },
    {
      "auxiliary_loss_clip": 0.01155017,
      "auxiliary_loss_mlp": 0.01023238,
      "balance_loss_clip": 1.04692388,
      "balance_loss_mlp": 1.01590645,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 4.585659660535567,
      "language_loss": 0.81284189,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83462447,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.644310712814331
    },
    {
      "auxiliary_loss_clip": 0.01161361,
      "auxiliary_loss_mlp": 0.0102386,
      "balance_loss_clip": 1.05480194,
      "balance_loss_mlp": 1.01672196,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 2.698377805918665,
      "language_loss": 0.87017918,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.89203137,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.632662057876587
    },
    {
      "auxiliary_loss_clip": 0.01152657,
      "auxiliary_loss_mlp": 0.01028615,
      "balance_loss_clip": 1.04747891,
      "balance_loss_mlp": 1.0203594,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 1.873365537661775,
      "language_loss": 0.7540679,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77588058,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.9325273036956787
    },
    {
      "auxiliary_loss_clip": 0.01079097,
      "auxiliary_loss_mlp": 0.01000965,
      "balance_loss_clip": 1.01932108,
      "balance_loss_mlp": 1.00000501,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.862519095082539,
      "language_loss": 0.60091841,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.621719,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.0375678539276123
    },
    {
      "auxiliary_loss_clip": 0.01151589,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.04653549,
      "balance_loss_mlp": 1.0186193,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 1.9911592785544965,
      "language_loss": 0.69069546,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.71247852,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.6367013454437256
    },
    {
      "auxiliary_loss_clip": 0.01175872,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.0509584,
      "balance_loss_mlp": 1.0162617,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 1.8408550816198672,
      "language_loss": 0.75049937,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77249396,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.666374444961548
    },
    {
      "auxiliary_loss_clip": 0.01171015,
      "auxiliary_loss_mlp": 0.0102334,
      "balance_loss_clip": 1.0491538,
      "balance_loss_mlp": 1.0163995,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 1.897951462060273,
      "language_loss": 0.72050762,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74245119,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.6429028511047363
    },
    {
      "auxiliary_loss_clip": 0.01058197,
      "auxiliary_loss_mlp": 0.01001018,
      "balance_loss_clip": 1.01951253,
      "balance_loss_mlp": 1.00000501,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.78451127158794,
      "language_loss": 0.53813505,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55872715,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 4.09476375579834
    },
    {
      "auxiliary_loss_clip": 0.01172636,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.0513016,
      "balance_loss_mlp": 1.01879215,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 1.9059044303197767,
      "language_loss": 0.81103694,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.8330214,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.663180112838745
    },
    {
      "auxiliary_loss_clip": 0.01169229,
      "auxiliary_loss_mlp": 0.01026269,
      "balance_loss_clip": 1.05235839,
      "balance_loss_mlp": 1.01869357,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 2.0507762253478288,
      "language_loss": 0.77669424,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.79864919,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.60842227935791
    },
    {
      "auxiliary_loss_clip": 0.01133569,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.04277849,
      "balance_loss_mlp": 1.01874411,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 2.391150003759821,
      "language_loss": 0.74152243,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76312196,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.764554262161255
    },
    {
      "auxiliary_loss_clip": 0.01115309,
      "auxiliary_loss_mlp": 0.01025356,
      "balance_loss_clip": 1.0422368,
      "balance_loss_mlp": 1.01819789,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 1.9588871461821953,
      "language_loss": 0.6764918,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69789839,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 3.7710230350494385
    },
    {
      "auxiliary_loss_clip": 0.01175351,
      "auxiliary_loss_mlp": 0.00900822,
      "balance_loss_clip": 1.0524987,
      "balance_loss_mlp": 1.00079727,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 2.3184399074673148,
      "language_loss": 0.74436152,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76512319,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.6134958267211914
    },
    {
      "auxiliary_loss_clip": 0.01136625,
      "auxiliary_loss_mlp": 0.01029397,
      "balance_loss_clip": 1.04855072,
      "balance_loss_mlp": 1.02177048,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.2340951990385434,
      "language_loss": 0.83819592,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85985613,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.701850175857544
    },
    {
      "auxiliary_loss_clip": 0.01160047,
      "auxiliary_loss_mlp": 0.01021449,
      "balance_loss_clip": 1.04856277,
      "balance_loss_mlp": 1.01410234,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.279699848395069,
      "language_loss": 0.8007502,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82256508,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.6189818382263184
    },
    {
      "auxiliary_loss_clip": 0.01168642,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.04917526,
      "balance_loss_mlp": 1.01591134,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.163445548860693,
      "language_loss": 0.79474199,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81666088,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.6797776222229004
    },
    {
      "auxiliary_loss_clip": 0.01164907,
      "auxiliary_loss_mlp": 0.00900194,
      "balance_loss_clip": 1.04838014,
      "balance_loss_mlp": 1.00078392,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 1.6973832111831828,
      "language_loss": 0.63145578,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65210682,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 3.681117534637451
    },
    {
      "auxiliary_loss_clip": 0.0114593,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.046296,
      "balance_loss_mlp": 1.01912808,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 1.9364778419695161,
      "language_loss": 0.8062706,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82799816,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.6784098148345947
    },
    {
      "auxiliary_loss_clip": 0.01154794,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.04654145,
      "balance_loss_mlp": 1.0182631,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.820955882586477,
      "language_loss": 0.7645843,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78639376,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.662271499633789
    },
    {
      "auxiliary_loss_clip": 0.01174164,
      "auxiliary_loss_mlp": 0.01020562,
      "balance_loss_clip": 1.05249596,
      "balance_loss_mlp": 1.01363277,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.5113769768583287,
      "language_loss": 0.73115891,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75310618,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.501427173614502
    },
    {
      "auxiliary_loss_clip": 0.01145627,
      "auxiliary_loss_mlp": 0.01028053,
      "balance_loss_clip": 1.04680514,
      "balance_loss_mlp": 1.02034664,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 1.929855457668682,
      "language_loss": 0.70666015,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72839695,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.7454206943511963
    },
    {
      "auxiliary_loss_clip": 0.01144873,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.04746568,
      "balance_loss_mlp": 1.01562405,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.0086328060052616,
      "language_loss": 0.85275483,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87443757,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.6620378494262695
    },
    {
      "auxiliary_loss_clip": 0.011541,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.04558778,
      "balance_loss_mlp": 1.01553094,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 2.0222867678176937,
      "language_loss": 0.68405306,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70582575,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.665658950805664
    },
    {
      "auxiliary_loss_clip": 0.01157765,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.04445028,
      "balance_loss_mlp": 1.02061296,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 2.6774546845164244,
      "language_loss": 0.70114875,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.7230112,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.653231143951416
    },
    {
      "auxiliary_loss_clip": 0.01165279,
      "auxiliary_loss_mlp": 0.01022082,
      "balance_loss_clip": 1.04864132,
      "balance_loss_mlp": 1.01485801,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 1.9991224259635594,
      "language_loss": 0.76973474,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.79160833,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.676386833190918
    },
    {
      "auxiliary_loss_clip": 0.0117645,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.05171704,
      "balance_loss_mlp": 1.02035916,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.9232297421763758,
      "language_loss": 0.75692242,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77896404,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 3.5264251232147217
    },
    {
      "auxiliary_loss_clip": 0.01170095,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.04990005,
      "balance_loss_mlp": 1.02228236,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 2.942994403327291,
      "language_loss": 0.62326127,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.645257,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.6266767978668213
    },
    {
      "auxiliary_loss_clip": 0.01146615,
      "auxiliary_loss_mlp": 0.01022632,
      "balance_loss_clip": 1.04555511,
      "balance_loss_mlp": 1.01512742,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.1334503082615326,
      "language_loss": 0.82862222,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.85031468,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.6956329345703125
    },
    {
      "auxiliary_loss_clip": 0.01165574,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.04951644,
      "balance_loss_mlp": 1.01811504,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.4275602384169006,
      "language_loss": 0.77255726,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79446769,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.589456081390381
    },
    {
      "auxiliary_loss_clip": 0.01168786,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 1.05156565,
      "balance_loss_mlp": 1.02331483,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 1.9049884286775918,
      "language_loss": 0.81044054,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83243984,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.611069917678833
    },
    {
      "auxiliary_loss_clip": 0.01147413,
      "auxiliary_loss_mlp": 0.01022363,
      "balance_loss_clip": 1.04606342,
      "balance_loss_mlp": 1.01498652,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.7827053320347976,
      "language_loss": 0.79372752,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81542528,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.7164556980133057
    },
    {
      "auxiliary_loss_clip": 0.01082744,
      "auxiliary_loss_mlp": 0.01001079,
      "balance_loss_clip": 1.01972485,
      "balance_loss_mlp": 1.00010765,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7634006079256187,
      "language_loss": 0.58303571,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60387385,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.2407617568969727
    },
    {
      "auxiliary_loss_clip": 0.01174433,
      "auxiliary_loss_mlp": 0.01023596,
      "balance_loss_clip": 1.05015385,
      "balance_loss_mlp": 1.01622927,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.657586525627747,
      "language_loss": 0.79704016,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81902051,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.5626134872436523
    },
    {
      "auxiliary_loss_clip": 0.01148356,
      "auxiliary_loss_mlp": 0.01022512,
      "balance_loss_clip": 1.04223502,
      "balance_loss_mlp": 1.01502228,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.8266894145255814,
      "language_loss": 0.85710335,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87881202,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.6338274478912354
    },
    {
      "auxiliary_loss_clip": 0.01153712,
      "auxiliary_loss_mlp": 0.01030901,
      "balance_loss_clip": 1.04267716,
      "balance_loss_mlp": 1.0225563,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 2.6980047898990778,
      "language_loss": 0.78493589,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80678201,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 3.682093381881714
    },
    {
      "auxiliary_loss_clip": 0.0115588,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.04747343,
      "balance_loss_mlp": 1.02043962,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.8072779561603902,
      "language_loss": 0.73663133,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.7584703,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.6293106079101562
    },
    {
      "auxiliary_loss_clip": 0.01165717,
      "auxiliary_loss_mlp": 0.01028707,
      "balance_loss_clip": 1.04878283,
      "balance_loss_mlp": 1.02122641,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.0841300707254544,
      "language_loss": 0.76317346,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78511769,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.706861734390259
    },
    {
      "auxiliary_loss_clip": 0.0114519,
      "auxiliary_loss_mlp": 0.01025513,
      "balance_loss_clip": 1.04812884,
      "balance_loss_mlp": 1.01812196,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.297779648810534,
      "language_loss": 0.75515062,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.77685761,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.6682119369506836
    },
    {
      "auxiliary_loss_clip": 0.01155862,
      "auxiliary_loss_mlp": 0.0102695,
      "balance_loss_clip": 1.0526439,
      "balance_loss_mlp": 1.01844501,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.6652763177736845,
      "language_loss": 0.74291855,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76474667,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 3.599682569503784
    },
    {
      "auxiliary_loss_clip": 0.01078281,
      "auxiliary_loss_mlp": 0.01001108,
      "balance_loss_clip": 1.01881218,
      "balance_loss_mlp": 1.00006485,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7405411201561182,
      "language_loss": 0.59550989,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.6163038,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.251398801803589
    },
    {
      "auxiliary_loss_clip": 0.01155558,
      "auxiliary_loss_mlp": 0.01024429,
      "balance_loss_clip": 1.04574347,
      "balance_loss_mlp": 1.01669216,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.4427146364475893,
      "language_loss": 0.70863187,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73043174,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.7430222034454346
    },
    {
      "auxiliary_loss_clip": 0.01174016,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.05138934,
      "balance_loss_mlp": 1.01867592,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.6587761638767942,
      "language_loss": 0.76059937,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78260225,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.6335830688476562
    },
    {
      "auxiliary_loss_clip": 0.01165216,
      "auxiliary_loss_mlp": 0.01023235,
      "balance_loss_clip": 1.05006909,
      "balance_loss_mlp": 1.01563287,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 2.0251621197573226,
      "language_loss": 0.7582674,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.78015196,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.5712928771972656
    },
    {
      "auxiliary_loss_clip": 0.0116311,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.05037403,
      "balance_loss_mlp": 1.01951897,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.2438428308975054,
      "language_loss": 0.75918627,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.78108847,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 3.567929983139038
    },
    {
      "auxiliary_loss_clip": 0.011576,
      "auxiliary_loss_mlp": 0.01021512,
      "balance_loss_clip": 1.0502516,
      "balance_loss_mlp": 1.01424289,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.9654269578889623,
      "language_loss": 0.66246039,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68425149,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.7149205207824707
    },
    {
      "auxiliary_loss_clip": 0.01149624,
      "auxiliary_loss_mlp": 0.01021868,
      "balance_loss_clip": 1.04450083,
      "balance_loss_mlp": 1.01484704,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.8739179009557974,
      "language_loss": 0.84801483,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.8697297,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.7241549491882324
    },
    {
      "auxiliary_loss_clip": 0.0115651,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 1.05194163,
      "balance_loss_mlp": 1.0229671,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.2566918121738264,
      "language_loss": 0.86151576,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88338733,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.5973780155181885
    },
    {
      "auxiliary_loss_clip": 0.01143518,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.04070342,
      "balance_loss_mlp": 1.0213567,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 3.0503422757280565,
      "language_loss": 0.74858391,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.77030867,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.8642139434814453
    },
    {
      "auxiliary_loss_clip": 0.011346,
      "auxiliary_loss_mlp": 0.00900513,
      "balance_loss_clip": 1.03993893,
      "balance_loss_mlp": 1.0008117,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 2.018630221067815,
      "language_loss": 0.80437362,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82472479,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.739380121231079
    },
    {
      "auxiliary_loss_clip": 0.01169212,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.04638863,
      "balance_loss_mlp": 1.0186336,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.8490175598112006,
      "language_loss": 0.69226813,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.71423119,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.6345582008361816
    },
    {
      "auxiliary_loss_clip": 0.01154615,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.05077183,
      "balance_loss_mlp": 1.01754832,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.9250773093647393,
      "language_loss": 0.73755014,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75934589,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.738147735595703
    },
    {
      "auxiliary_loss_clip": 0.01168791,
      "auxiliary_loss_mlp": 0.01022885,
      "balance_loss_clip": 1.05099523,
      "balance_loss_mlp": 1.0159235,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.509697313583302,
      "language_loss": 0.80897033,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.83088708,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.6455256938934326
    },
    {
      "auxiliary_loss_clip": 0.01130413,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.04290164,
      "balance_loss_mlp": 1.0197165,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.8976943778339976,
      "language_loss": 0.72240317,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74397749,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 3.778850793838501
    },
    {
      "auxiliary_loss_clip": 0.01159512,
      "auxiliary_loss_mlp": 0.01022311,
      "balance_loss_clip": 1.04865754,
      "balance_loss_mlp": 1.0149436,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.5615464607718017,
      "language_loss": 0.68464828,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70646656,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.643386125564575
    },
    {
      "auxiliary_loss_clip": 0.01156834,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.04573131,
      "balance_loss_mlp": 1.019382,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 1.9763008441335348,
      "language_loss": 0.80849314,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83033746,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.68589186668396
    },
    {
      "auxiliary_loss_clip": 0.01151681,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04566014,
      "balance_loss_mlp": 1.01747036,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 2.247896313592386,
      "language_loss": 0.84208828,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86385232,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.6579108238220215
    },
    {
      "auxiliary_loss_clip": 0.0116025,
      "auxiliary_loss_mlp": 0.01026857,
      "balance_loss_clip": 1.04540849,
      "balance_loss_mlp": 1.0191679,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.1152880094278164,
      "language_loss": 0.71984899,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74172002,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.6346218585968018
    },
    {
      "auxiliary_loss_clip": 0.01066875,
      "auxiliary_loss_mlp": 0.00890979,
      "balance_loss_clip": 1.02144885,
      "balance_loss_mlp": 1.00002098,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9258745187256694,
      "language_loss": 0.64408171,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66366029,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.079331874847412
    },
    {
      "auxiliary_loss_clip": 0.01066834,
      "auxiliary_loss_mlp": 0.00890777,
      "balance_loss_clip": 1.01919508,
      "balance_loss_mlp": 0.99998707,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.1047425824800698,
      "language_loss": 0.59657669,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61615282,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.097228765487671
    },
    {
      "auxiliary_loss_clip": 0.01149858,
      "auxiliary_loss_mlp": 0.01028437,
      "balance_loss_clip": 1.04448318,
      "balance_loss_mlp": 1.02108467,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.782999841584221,
      "language_loss": 0.76318884,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78497171,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.7132773399353027
    },
    {
      "auxiliary_loss_clip": 0.01112454,
      "auxiliary_loss_mlp": 0.01025117,
      "balance_loss_clip": 1.04203987,
      "balance_loss_mlp": 1.01817346,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 1.8979080499117118,
      "language_loss": 0.69118845,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71256411,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.838873863220215
    },
    {
      "auxiliary_loss_clip": 0.01166726,
      "auxiliary_loss_mlp": 0.00900734,
      "balance_loss_clip": 1.05049968,
      "balance_loss_mlp": 1.00089073,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 1.8012240217787794,
      "language_loss": 0.75378191,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77445662,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 3.8803000450134277
    },
    {
      "auxiliary_loss_clip": 0.01154737,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04928982,
      "balance_loss_mlp": 1.02040648,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.920886056157598,
      "language_loss": 0.86467361,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.88650012,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.711181640625
    },
    {
      "auxiliary_loss_clip": 0.01137149,
      "auxiliary_loss_mlp": 0.01027998,
      "balance_loss_clip": 1.04404116,
      "balance_loss_mlp": 1.02029705,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.8805669604616861,
      "language_loss": 0.80338788,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82503927,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.6946444511413574
    },
    {
      "auxiliary_loss_clip": 0.01160936,
      "auxiliary_loss_mlp": 0.01028592,
      "balance_loss_clip": 1.04777575,
      "balance_loss_mlp": 1.02120662,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.8720273659270676,
      "language_loss": 0.80944264,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.83133793,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.753937244415283
    },
    {
      "auxiliary_loss_clip": 0.01151329,
      "auxiliary_loss_mlp": 0.01025757,
      "balance_loss_clip": 1.04662681,
      "balance_loss_mlp": 1.01816893,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.8039261088441298,
      "language_loss": 0.8077904,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.82956123,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 3.665431499481201
    },
    {
      "auxiliary_loss_clip": 0.01147041,
      "auxiliary_loss_mlp": 0.01022065,
      "balance_loss_clip": 1.04449534,
      "balance_loss_mlp": 1.01495755,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.8941455278420039,
      "language_loss": 0.86512065,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88681179,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.6967732906341553
    },
    {
      "auxiliary_loss_clip": 0.01153725,
      "auxiliary_loss_mlp": 0.0102156,
      "balance_loss_clip": 1.04465532,
      "balance_loss_mlp": 1.01392198,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.7897240491938085,
      "language_loss": 0.85761613,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87936902,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.7408313751220703
    },
    {
      "auxiliary_loss_clip": 0.01173479,
      "auxiliary_loss_mlp": 0.00901203,
      "balance_loss_clip": 1.04812002,
      "balance_loss_mlp": 1.00084126,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.8061834091497881,
      "language_loss": 0.76526731,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78601414,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.5760412216186523
    },
    {
      "auxiliary_loss_clip": 0.01161035,
      "auxiliary_loss_mlp": 0.01031145,
      "balance_loss_clip": 1.04819334,
      "balance_loss_mlp": 1.02325308,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.9112105801232597,
      "language_loss": 0.866552,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88847375,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 3.67382550239563
    },
    {
      "auxiliary_loss_clip": 0.01166401,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.05192626,
      "balance_loss_mlp": 1.02084243,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.168674315408315,
      "language_loss": 0.72243935,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74439013,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.5873212814331055
    },
    {
      "auxiliary_loss_clip": 0.01153647,
      "auxiliary_loss_mlp": 0.01027468,
      "balance_loss_clip": 1.04825747,
      "balance_loss_mlp": 1.01961184,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 3.4933441344757896,
      "language_loss": 0.49018571,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51199681,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.6383824348449707
    },
    {
      "auxiliary_loss_clip": 0.01154326,
      "auxiliary_loss_mlp": 0.0102746,
      "balance_loss_clip": 1.04889894,
      "balance_loss_mlp": 1.02000344,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.9192382735301383,
      "language_loss": 0.56384385,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58566171,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.683925151824951
    },
    {
      "auxiliary_loss_clip": 0.01156653,
      "auxiliary_loss_mlp": 0.01022871,
      "balance_loss_clip": 1.04770839,
      "balance_loss_mlp": 1.01545596,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 1.9039432242726637,
      "language_loss": 0.65725386,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67904913,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.64758563041687
    },
    {
      "auxiliary_loss_clip": 0.01167268,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.05050755,
      "balance_loss_mlp": 1.02189028,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.4793747703658506,
      "language_loss": 0.72038841,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74235636,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.6611692905426025
    },
    {
      "auxiliary_loss_clip": 0.01141947,
      "auxiliary_loss_mlp": 0.01026898,
      "balance_loss_clip": 1.0462265,
      "balance_loss_mlp": 1.01880407,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.6073277087564175,
      "language_loss": 0.78131223,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80300069,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.7288737297058105
    },
    {
      "auxiliary_loss_clip": 0.01154411,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.04768062,
      "balance_loss_mlp": 1.01790118,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 2.2449306946575858,
      "language_loss": 0.74208522,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.7638824,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.6674869060516357
    },
    {
      "auxiliary_loss_clip": 0.01137447,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.04041839,
      "balance_loss_mlp": 1.01872516,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 2.1266127958493235,
      "language_loss": 0.73631024,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75794947,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.8162219524383545
    },
    {
      "auxiliary_loss_clip": 0.0115863,
      "auxiliary_loss_mlp": 0.01027823,
      "balance_loss_clip": 1.04552245,
      "balance_loss_mlp": 1.02006853,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 2.110893440467739,
      "language_loss": 0.7027986,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72466314,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 3.6035890579223633
    },
    {
      "auxiliary_loss_clip": 0.01168695,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.05089021,
      "balance_loss_mlp": 1.02050328,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.347325543256494,
      "language_loss": 0.70595765,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72792953,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.564455986022949
    },
    {
      "auxiliary_loss_clip": 0.01068883,
      "auxiliary_loss_mlp": 0.01011591,
      "balance_loss_clip": 1.02821934,
      "balance_loss_mlp": 1.01042831,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.8373600992223661,
      "language_loss": 0.60418499,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62498975,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.2041945457458496
    },
    {
      "auxiliary_loss_clip": 0.01176482,
      "auxiliary_loss_mlp": 0.01018853,
      "balance_loss_clip": 1.05157781,
      "balance_loss_mlp": 1.01140523,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 1.7129881213702216,
      "language_loss": 0.82241607,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84436941,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.657729148864746
    },
    {
      "auxiliary_loss_clip": 0.01166572,
      "auxiliary_loss_mlp": 0.01019591,
      "balance_loss_clip": 1.04674101,
      "balance_loss_mlp": 1.01193476,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.8415507738424817,
      "language_loss": 0.83400393,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85586554,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.6151680946350098
    },
    {
      "auxiliary_loss_clip": 0.01068823,
      "auxiliary_loss_mlp": 0.01009673,
      "balance_loss_clip": 1.02901542,
      "balance_loss_mlp": 1.00863636,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8232062488015072,
      "language_loss": 0.580984,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60176897,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.2945456504821777
    },
    {
      "auxiliary_loss_clip": 0.01173019,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 1.05192137,
      "balance_loss_mlp": 1.0159359,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 1.973394209810477,
      "language_loss": 0.7692337,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.79119545,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 2.527167797088623
    },
    {
      "auxiliary_loss_clip": 0.01145899,
      "auxiliary_loss_mlp": 0.00901521,
      "balance_loss_clip": 1.04202509,
      "balance_loss_mlp": 1.00074649,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.8159601102734406,
      "language_loss": 0.82310951,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.8435837,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.6994924545288086
    },
    {
      "auxiliary_loss_clip": 0.01168566,
      "auxiliary_loss_mlp": 0.01024394,
      "balance_loss_clip": 1.05013287,
      "balance_loss_mlp": 1.0169909,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.6813927631810235,
      "language_loss": 0.70380223,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72573185,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.623577356338501
    },
    {
      "auxiliary_loss_clip": 0.01152013,
      "auxiliary_loss_mlp": 0.01032358,
      "balance_loss_clip": 1.04886746,
      "balance_loss_mlp": 1.02491963,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.9674009781780646,
      "language_loss": 0.77672303,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79856676,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 3.6592507362365723
    },
    {
      "auxiliary_loss_clip": 0.01146852,
      "auxiliary_loss_mlp": 0.01025567,
      "balance_loss_clip": 1.04597354,
      "balance_loss_mlp": 1.01776457,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 1.9931021416752035,
      "language_loss": 0.81752729,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83925152,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.7421987056732178
    },
    {
      "auxiliary_loss_clip": 0.0116775,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.0517534,
      "balance_loss_mlp": 1.02391696,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 2.000342313261341,
      "language_loss": 0.88533568,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90733159,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.556225299835205
    },
    {
      "auxiliary_loss_clip": 0.01137378,
      "auxiliary_loss_mlp": 0.01022365,
      "balance_loss_clip": 1.04238796,
      "balance_loss_mlp": 1.01529288,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.38819165681666,
      "language_loss": 0.72987252,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.75146997,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.6574511528015137
    },
    {
      "auxiliary_loss_clip": 0.01155273,
      "auxiliary_loss_mlp": 0.01027568,
      "balance_loss_clip": 1.04464936,
      "balance_loss_mlp": 1.01969457,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 2.1036857652630596,
      "language_loss": 0.62902439,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.65085286,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 3.539621353149414
    },
    {
      "auxiliary_loss_clip": 0.01148994,
      "auxiliary_loss_mlp": 0.01030907,
      "balance_loss_clip": 1.04687452,
      "balance_loss_mlp": 1.02319419,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.1946201988818594,
      "language_loss": 0.76505554,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78685451,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.699585437774658
    },
    {
      "auxiliary_loss_clip": 0.01173694,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.05095732,
      "balance_loss_mlp": 1.02053189,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.985848727421763,
      "language_loss": 0.87054348,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89256722,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.584981918334961
    },
    {
      "auxiliary_loss_clip": 0.01173806,
      "auxiliary_loss_mlp": 0.01026726,
      "balance_loss_clip": 1.05115509,
      "balance_loss_mlp": 1.01953769,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 1.7946644459450098,
      "language_loss": 0.68086618,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.70287156,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.6592414379119873
    },
    {
      "auxiliary_loss_clip": 0.01145252,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.04541981,
      "balance_loss_mlp": 1.01884413,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.4857720792836004,
      "language_loss": 0.71795261,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.73967439,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.770972967147827
    },
    {
      "auxiliary_loss_clip": 0.01167219,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.05067039,
      "balance_loss_mlp": 1.02141273,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.1046052188793554,
      "language_loss": 0.71108341,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73305959,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 3.942261219024658
    },
    {
      "auxiliary_loss_clip": 0.01153493,
      "auxiliary_loss_mlp": 0.00901574,
      "balance_loss_clip": 1.04516709,
      "balance_loss_mlp": 1.0008049,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 2.00028516242809,
      "language_loss": 0.81655657,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.83710718,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.726508140563965
    },
    {
      "auxiliary_loss_clip": 0.01154688,
      "auxiliary_loss_mlp": 0.01025901,
      "balance_loss_clip": 1.04690969,
      "balance_loss_mlp": 1.01873064,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 2.410475640298611,
      "language_loss": 0.68879312,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.71059901,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.601469039916992
    },
    {
      "auxiliary_loss_clip": 0.01079588,
      "auxiliary_loss_mlp": 0.01003795,
      "balance_loss_clip": 1.02066016,
      "balance_loss_mlp": 1.00287151,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7378208110856922,
      "language_loss": 0.58056104,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60139489,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.250563383102417
    },
    {
      "auxiliary_loss_clip": 0.01150643,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04639506,
      "balance_loss_mlp": 1.01691604,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 2.7504414812837146,
      "language_loss": 0.78543675,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80719721,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.6562466621398926
    },
    {
      "auxiliary_loss_clip": 0.01155279,
      "auxiliary_loss_mlp": 0.0103393,
      "balance_loss_clip": 1.04937577,
      "balance_loss_mlp": 1.02606273,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.7769316542560782,
      "language_loss": 0.80482781,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82671988,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.6498072147369385
    },
    {
      "auxiliary_loss_clip": 0.01143556,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.0448482,
      "balance_loss_mlp": 1.01763248,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 1.892959870115667,
      "language_loss": 0.79951274,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82119787,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.743934392929077
    },
    {
      "auxiliary_loss_clip": 0.01138772,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.0434587,
      "balance_loss_mlp": 1.01739991,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.628176998834996,
      "language_loss": 0.75787461,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.77950716,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.6585283279418945
    },
    {
      "auxiliary_loss_clip": 0.01107091,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.03839731,
      "balance_loss_mlp": 1.02334201,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.9434059923763871,
      "language_loss": 0.75067896,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77206105,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.8745574951171875
    },
    {
      "auxiliary_loss_clip": 0.01176305,
      "auxiliary_loss_mlp": 0.01029943,
      "balance_loss_clip": 1.05080628,
      "balance_loss_mlp": 1.0221231,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.1054359064011794,
      "language_loss": 0.78019869,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80226117,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.8206658363342285
    },
    {
      "auxiliary_loss_clip": 0.01143561,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.04611814,
      "balance_loss_mlp": 1.01822734,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 2.7315563718992895,
      "language_loss": 0.72947329,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75116295,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 3.6390488147735596
    },
    {
      "auxiliary_loss_clip": 0.01148375,
      "auxiliary_loss_mlp": 0.01028594,
      "balance_loss_clip": 1.04469776,
      "balance_loss_mlp": 1.02039802,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.142250944006934,
      "language_loss": 0.66487134,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.68664104,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.802842378616333
    },
    {
      "auxiliary_loss_clip": 0.01152557,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 1.04685307,
      "balance_loss_mlp": 1.02216041,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.6612262197141123,
      "language_loss": 0.72126764,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74308819,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.697528600692749
    },
    {
      "auxiliary_loss_clip": 0.01167279,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.05159283,
      "balance_loss_mlp": 1.01962996,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 3.386948775948553,
      "language_loss": 0.74911839,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.77106041,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.6342525482177734
    },
    {
      "auxiliary_loss_clip": 0.0116357,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.04841137,
      "balance_loss_mlp": 1.02289724,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 2.300030341874095,
      "language_loss": 0.77455938,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79650033,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.6622262001037598
    },
    {
      "auxiliary_loss_clip": 0.01144013,
      "auxiliary_loss_mlp": 0.01024933,
      "balance_loss_clip": 1.04754257,
      "balance_loss_mlp": 1.0170058,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 3.617132123855141,
      "language_loss": 0.59642917,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61811864,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.769571304321289
    },
    {
      "auxiliary_loss_clip": 0.01160471,
      "auxiliary_loss_mlp": 0.0102832,
      "balance_loss_clip": 1.04708743,
      "balance_loss_mlp": 1.02057099,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.0946366176358415,
      "language_loss": 0.82417673,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84606463,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.62174129486084
    },
    {
      "auxiliary_loss_clip": 0.01160622,
      "auxiliary_loss_mlp": 0.01026418,
      "balance_loss_clip": 1.0500493,
      "balance_loss_mlp": 1.01917005,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.1709284179784647,
      "language_loss": 0.6812523,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70312274,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.656248092651367
    },
    {
      "auxiliary_loss_clip": 0.01151073,
      "auxiliary_loss_mlp": 0.00901274,
      "balance_loss_clip": 1.04776442,
      "balance_loss_mlp": 1.00086069,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.9164503329211184,
      "language_loss": 0.74291885,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76344234,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 3.878648042678833
    },
    {
      "auxiliary_loss_clip": 0.01158116,
      "auxiliary_loss_mlp": 0.01020845,
      "balance_loss_clip": 1.04448223,
      "balance_loss_mlp": 1.0131737,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.6898361873635466,
      "language_loss": 0.68043345,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70222312,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.7957170009613037
    },
    {
      "auxiliary_loss_clip": 0.01079101,
      "auxiliary_loss_mlp": 0.01004189,
      "balance_loss_clip": 1.02875388,
      "balance_loss_mlp": 1.00305068,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8496437475743885,
      "language_loss": 0.65042156,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67125446,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.222412109375
    },
    {
      "auxiliary_loss_clip": 0.01145136,
      "auxiliary_loss_mlp": 0.00901028,
      "balance_loss_clip": 1.05087996,
      "balance_loss_mlp": 1.00082338,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 2.1717735020453466,
      "language_loss": 0.75130856,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.77177018,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.6153969764709473
    },
    {
      "auxiliary_loss_clip": 0.01155653,
      "auxiliary_loss_mlp": 0.0102015,
      "balance_loss_clip": 1.04853058,
      "balance_loss_mlp": 1.01285422,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.6034664978521556,
      "language_loss": 0.83166492,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.853423,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 3.631700277328491
    },
    {
      "auxiliary_loss_clip": 0.01140863,
      "auxiliary_loss_mlp": 0.0102308,
      "balance_loss_clip": 1.04617691,
      "balance_loss_mlp": 1.01618922,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.73943855078028,
      "language_loss": 0.72041821,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74205768,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.828688859939575
    },
    {
      "auxiliary_loss_clip": 0.0115099,
      "auxiliary_loss_mlp": 0.0102582,
      "balance_loss_clip": 1.0463326,
      "balance_loss_mlp": 1.01805663,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.692408535422642,
      "language_loss": 0.73725188,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75901997,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.7086052894592285
    },
    {
      "auxiliary_loss_clip": 0.0116759,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.05159903,
      "balance_loss_mlp": 1.01692247,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.184083075994519,
      "language_loss": 0.7075761,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72950065,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.572451591491699
    },
    {
      "auxiliary_loss_clip": 0.01169735,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.05089021,
      "balance_loss_mlp": 1.01853538,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 2.039315698972509,
      "language_loss": 0.73987186,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76183277,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 3.54437255859375
    },
    {
      "auxiliary_loss_clip": 0.01128557,
      "auxiliary_loss_mlp": 0.01033672,
      "balance_loss_clip": 1.04210901,
      "balance_loss_mlp": 1.02543461,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.8271040509040528,
      "language_loss": 0.73831201,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75993431,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.700805902481079
    },
    {
      "auxiliary_loss_clip": 0.01151068,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04418111,
      "balance_loss_mlp": 1.01753414,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.4329890640528884,
      "language_loss": 0.62061644,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.64238465,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.6344799995422363
    },
    {
      "auxiliary_loss_clip": 0.0105471,
      "auxiliary_loss_mlp": 0.01003054,
      "balance_loss_clip": 1.01875257,
      "balance_loss_mlp": 1.0020411,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.7975082884338401,
      "language_loss": 0.57378989,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59436756,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.1989803314208984
    },
    {
      "auxiliary_loss_clip": 0.01158418,
      "auxiliary_loss_mlp": 0.0090105,
      "balance_loss_clip": 1.04710913,
      "balance_loss_mlp": 1.00076747,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 1.9050182125836927,
      "language_loss": 0.78332746,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80392206,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.7465388774871826
    },
    {
      "auxiliary_loss_clip": 0.01170758,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.050349,
      "balance_loss_mlp": 1.01681185,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.8839413486581174,
      "language_loss": 0.61298645,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63494456,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.597360610961914
    },
    {
      "auxiliary_loss_clip": 0.0114795,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.05010736,
      "balance_loss_mlp": 1.01994407,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 2.0715430415945915,
      "language_loss": 0.77458984,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.79635215,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.7448315620422363
    },
    {
      "auxiliary_loss_clip": 0.01158061,
      "auxiliary_loss_mlp": 0.0102296,
      "balance_loss_clip": 1.04532588,
      "balance_loss_mlp": 1.01534796,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.9584962772995456,
      "language_loss": 0.85190481,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87371504,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.5787651538848877
    },
    {
      "auxiliary_loss_clip": 0.01108662,
      "auxiliary_loss_mlp": 0.01031092,
      "balance_loss_clip": 1.03895617,
      "balance_loss_mlp": 1.02219951,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 2.3628629742597194,
      "language_loss": 0.68737733,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70877492,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.904212474822998
    },
    {
      "auxiliary_loss_clip": 0.01153924,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.04903698,
      "balance_loss_mlp": 1.01952147,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.7171304669429621,
      "language_loss": 0.72042799,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74223936,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.834632635116577
    },
    {
      "auxiliary_loss_clip": 0.0114772,
      "auxiliary_loss_mlp": 0.01024678,
      "balance_loss_clip": 1.04644299,
      "balance_loss_mlp": 1.01761174,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 3.376143274610986,
      "language_loss": 0.82226169,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84398568,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 3.605786085128784
    },
    {
      "auxiliary_loss_clip": 0.0114707,
      "auxiliary_loss_mlp": 0.00901612,
      "balance_loss_clip": 1.04386866,
      "balance_loss_mlp": 1.00082898,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 1.88973578028138,
      "language_loss": 0.90173995,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92222679,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.693742036819458
    },
    {
      "auxiliary_loss_clip": 0.01153595,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.04533458,
      "balance_loss_mlp": 1.01972222,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.747132285178582,
      "language_loss": 0.7227717,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74458015,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.6478402614593506
    },
    {
      "auxiliary_loss_clip": 0.01163907,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.0467906,
      "balance_loss_mlp": 1.01826239,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 1.845060301295336,
      "language_loss": 0.77158868,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79348171,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.6875884532928467
    },
    {
      "auxiliary_loss_clip": 0.01165911,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.04816914,
      "balance_loss_mlp": 1.0181421,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 2.6847214395116974,
      "language_loss": 0.85194659,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87385297,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.545637845993042
    },
    {
      "auxiliary_loss_clip": 0.01144494,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.04516149,
      "balance_loss_mlp": 1.01947129,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 2.6104603302180576,
      "language_loss": 0.78257352,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80428576,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.751049518585205
    },
    {
      "auxiliary_loss_clip": 0.01147884,
      "auxiliary_loss_mlp": 0.01023904,
      "balance_loss_clip": 1.04907739,
      "balance_loss_mlp": 1.01646841,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.7986526503370182,
      "language_loss": 0.8105824,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83230025,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 2.70507550239563
    },
    {
      "auxiliary_loss_clip": 0.01176286,
      "auxiliary_loss_mlp": 0.01029595,
      "balance_loss_clip": 1.05126882,
      "balance_loss_mlp": 1.02178657,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.4156702168095894,
      "language_loss": 0.778579,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.80063778,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.629617691040039
    },
    {
      "auxiliary_loss_clip": 0.01150095,
      "auxiliary_loss_mlp": 0.01028267,
      "balance_loss_clip": 1.04738438,
      "balance_loss_mlp": 1.02055967,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 3.9753053571258863,
      "language_loss": 0.87602544,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89780903,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 3.77185320854187
    },
    {
      "auxiliary_loss_clip": 0.01149416,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04538918,
      "balance_loss_mlp": 1.01962912,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.8740235188997876,
      "language_loss": 0.88633192,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90810156,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.775458574295044
    },
    {
      "auxiliary_loss_clip": 0.01157652,
      "auxiliary_loss_mlp": 0.01031565,
      "balance_loss_clip": 1.04691589,
      "balance_loss_mlp": 1.0234108,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 2.0511410141001165,
      "language_loss": 0.72641277,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.74830484,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.6358702182769775
    },
    {
      "auxiliary_loss_clip": 0.01155836,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.04840338,
      "balance_loss_mlp": 1.01910734,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 5.014256139473835,
      "language_loss": 0.67527163,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.69709527,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.6063435077667236
    },
    {
      "auxiliary_loss_clip": 0.01137089,
      "auxiliary_loss_mlp": 0.01028655,
      "balance_loss_clip": 1.04558134,
      "balance_loss_mlp": 1.02107954,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.6836495225284562,
      "language_loss": 0.74834681,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.77000427,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 3.5976102352142334
    },
    {
      "auxiliary_loss_clip": 0.01131877,
      "auxiliary_loss_mlp": 0.01028188,
      "balance_loss_clip": 1.04391956,
      "balance_loss_mlp": 1.02070761,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.6513872837916856,
      "language_loss": 0.75287271,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77447337,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.7612991333007812
    },
    {
      "auxiliary_loss_clip": 0.01150157,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.04451692,
      "balance_loss_mlp": 1.02182424,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 2.2027866019744016,
      "language_loss": 0.74914694,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.77094799,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.639270067214966
    },
    {
      "auxiliary_loss_clip": 0.01172567,
      "auxiliary_loss_mlp": 0.01023533,
      "balance_loss_clip": 1.04893517,
      "balance_loss_mlp": 1.01618958,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 1.766898761754994,
      "language_loss": 0.72908008,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75104105,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.555016279220581
    },
    {
      "auxiliary_loss_clip": 0.01175757,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.05235672,
      "balance_loss_mlp": 1.02010643,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.5275847052555913,
      "language_loss": 0.72059858,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74264348,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.6293599605560303
    },
    {
      "auxiliary_loss_clip": 0.01174637,
      "auxiliary_loss_mlp": 0.01024773,
      "balance_loss_clip": 1.04999912,
      "balance_loss_mlp": 1.01676202,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.7729617948589564,
      "language_loss": 0.69087243,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71286654,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 3.6617496013641357
    },
    {
      "auxiliary_loss_clip": 0.01177155,
      "auxiliary_loss_mlp": 0.01032179,
      "balance_loss_clip": 1.05006635,
      "balance_loss_mlp": 1.02420402,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.9022629937433733,
      "language_loss": 0.71739954,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.73949283,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.5631823539733887
    },
    {
      "auxiliary_loss_clip": 0.01176526,
      "auxiliary_loss_mlp": 0.00901464,
      "balance_loss_clip": 1.05071473,
      "balance_loss_mlp": 1.00080895,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.1820616150607126,
      "language_loss": 0.76821011,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.78899002,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.5564332008361816
    },
    {
      "auxiliary_loss_clip": 0.01161428,
      "auxiliary_loss_mlp": 0.00902178,
      "balance_loss_clip": 1.04697156,
      "balance_loss_mlp": 1.00074625,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.7960352239563835,
      "language_loss": 0.693097,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.71373302,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.587402820587158
    },
    {
      "auxiliary_loss_clip": 0.01161743,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04745793,
      "balance_loss_mlp": 1.01945019,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 2.3373143076819907,
      "language_loss": 0.79780698,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.81969756,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.6083221435546875
    },
    {
      "auxiliary_loss_clip": 0.01172329,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.04916716,
      "balance_loss_mlp": 1.02067614,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 1.6800332888613176,
      "language_loss": 0.77303672,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79504687,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.5294854640960693
    },
    {
      "auxiliary_loss_clip": 0.01170316,
      "auxiliary_loss_mlp": 0.00901866,
      "balance_loss_clip": 1.05242205,
      "balance_loss_mlp": 1.00070548,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 4.695762693752064,
      "language_loss": 0.68264902,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70337081,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.6706435680389404
    },
    {
      "auxiliary_loss_clip": 0.01142038,
      "auxiliary_loss_mlp": 0.01029234,
      "balance_loss_clip": 1.04293799,
      "balance_loss_mlp": 1.02099717,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 1.9701408710096455,
      "language_loss": 0.77435672,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79606944,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.669499635696411
    },
    {
      "auxiliary_loss_clip": 0.01156492,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.046561,
      "balance_loss_mlp": 1.01782572,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.68424965454219,
      "language_loss": 0.7456615,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76747787,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.718122959136963
    },
    {
      "auxiliary_loss_clip": 0.01157393,
      "auxiliary_loss_mlp": 0.01031212,
      "balance_loss_clip": 1.04516435,
      "balance_loss_mlp": 1.02330244,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 4.5290838170150005,
      "language_loss": 0.85224545,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.8741315,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 3.563596725463867
    },
    {
      "auxiliary_loss_clip": 0.01173289,
      "auxiliary_loss_mlp": 0.01023882,
      "balance_loss_clip": 1.05087399,
      "balance_loss_mlp": 1.016348,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.7455264966267183,
      "language_loss": 0.82632881,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84830046,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.553586006164551
    },
    {
      "auxiliary_loss_clip": 0.01171033,
      "auxiliary_loss_mlp": 0.01026796,
      "balance_loss_clip": 1.04933679,
      "balance_loss_mlp": 1.01922584,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 4.793620927838953,
      "language_loss": 0.62107968,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64305806,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.654876232147217
    },
    {
      "auxiliary_loss_clip": 0.01162592,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.04940367,
      "balance_loss_mlp": 1.02257609,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 2.0008772522140137,
      "language_loss": 0.80906308,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83098745,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.754481792449951
    },
    {
      "auxiliary_loss_clip": 0.01163792,
      "auxiliary_loss_mlp": 0.01024939,
      "balance_loss_clip": 1.0458082,
      "balance_loss_mlp": 1.01744103,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.8318460999613053,
      "language_loss": 0.78056139,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80244863,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 2.6073594093322754
    },
    {
      "auxiliary_loss_clip": 0.01166622,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.04631138,
      "balance_loss_mlp": 1.0166297,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 3.1948442530362056,
      "language_loss": 0.88428986,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90620589,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.550137519836426
    },
    {
      "auxiliary_loss_clip": 0.01150572,
      "auxiliary_loss_mlp": 0.01024464,
      "balance_loss_clip": 1.04806709,
      "balance_loss_mlp": 1.01737118,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 1.7868427077261675,
      "language_loss": 0.72067416,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74242449,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.734238862991333
    },
    {
      "auxiliary_loss_clip": 0.01171094,
      "auxiliary_loss_mlp": 0.01028524,
      "balance_loss_clip": 1.05080295,
      "balance_loss_mlp": 1.02099621,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.57661228435777,
      "language_loss": 0.62405282,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64604902,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.5641608238220215
    },
    {
      "auxiliary_loss_clip": 0.01162601,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04609239,
      "balance_loss_mlp": 1.01795626,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.7433654107585927,
      "language_loss": 0.74440354,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.76628089,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.6676933765411377
    },
    {
      "auxiliary_loss_clip": 0.01148224,
      "auxiliary_loss_mlp": 0.01029628,
      "balance_loss_clip": 1.041713,
      "balance_loss_mlp": 1.02218318,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 2.2213372352982463,
      "language_loss": 0.66489738,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68667585,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 3.8203773498535156
    },
    {
      "auxiliary_loss_clip": 0.01169796,
      "auxiliary_loss_mlp": 0.01034283,
      "balance_loss_clip": 1.05276859,
      "balance_loss_mlp": 1.02540135,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 3.0529782409189163,
      "language_loss": 0.70633543,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72837627,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.7531285285949707
    },
    {
      "auxiliary_loss_clip": 0.01158632,
      "auxiliary_loss_mlp": 0.01027083,
      "balance_loss_clip": 1.04735112,
      "balance_loss_mlp": 1.01964474,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 2.349414200141234,
      "language_loss": 0.62482417,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64668131,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.883052110671997
    },
    {
      "auxiliary_loss_clip": 0.01137068,
      "auxiliary_loss_mlp": 0.01029765,
      "balance_loss_clip": 1.04478908,
      "balance_loss_mlp": 1.02130759,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 1.8487700943842547,
      "language_loss": 0.70765722,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.72932559,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.6631579399108887
    },
    {
      "auxiliary_loss_clip": 0.01131214,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.04324961,
      "balance_loss_mlp": 1.0181756,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.3039944409143316,
      "language_loss": 0.83643687,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85801321,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 3.5903899669647217
    },
    {
      "auxiliary_loss_clip": 0.0115833,
      "auxiliary_loss_mlp": 0.01026711,
      "balance_loss_clip": 1.04826665,
      "balance_loss_mlp": 1.01895678,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 2.024391366709723,
      "language_loss": 0.90151882,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92336923,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.760631561279297
    },
    {
      "auxiliary_loss_clip": 0.01175648,
      "auxiliary_loss_mlp": 0.01023465,
      "balance_loss_clip": 1.05155396,
      "balance_loss_mlp": 1.01619315,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.9594520566198939,
      "language_loss": 0.76751018,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.78950131,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.5632212162017822
    },
    {
      "auxiliary_loss_clip": 0.01150557,
      "auxiliary_loss_mlp": 0.01027508,
      "balance_loss_clip": 1.04790306,
      "balance_loss_mlp": 1.01892543,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5345696207536597,
      "language_loss": 0.72923219,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.75101286,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.787930727005005
    },
    {
      "auxiliary_loss_clip": 0.01164771,
      "auxiliary_loss_mlp": 0.01027237,
      "balance_loss_clip": 1.04963899,
      "balance_loss_mlp": 1.01958346,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 2.269062165357058,
      "language_loss": 0.75094694,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77286702,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 3.461422920227051
    },
    {
      "auxiliary_loss_clip": 0.01138192,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04852319,
      "balance_loss_mlp": 1.018363,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 2.212134129265603,
      "language_loss": 0.69496417,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71660829,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.774401903152466
    },
    {
      "auxiliary_loss_clip": 0.01164531,
      "auxiliary_loss_mlp": 0.0102509,
      "balance_loss_clip": 1.04633212,
      "balance_loss_mlp": 1.01750207,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.885292365596002,
      "language_loss": 0.77905035,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.80094659,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.574777603149414
    },
    {
      "auxiliary_loss_clip": 0.01142548,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.05052423,
      "balance_loss_mlp": 1.0169872,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 2.344297835693254,
      "language_loss": 0.78720498,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.808873,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.6902518272399902
    },
    {
      "auxiliary_loss_clip": 0.01176416,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.05208182,
      "balance_loss_mlp": 1.02168036,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 2.0393817034172637,
      "language_loss": 0.78616762,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80822116,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.6483652591705322
    },
    {
      "auxiliary_loss_clip": 0.01142813,
      "auxiliary_loss_mlp": 0.01028583,
      "balance_loss_clip": 1.045048,
      "balance_loss_mlp": 1.0212841,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 2.1679551360237763,
      "language_loss": 0.77284318,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79455715,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.7528347969055176
    },
    {
      "auxiliary_loss_clip": 0.01163463,
      "auxiliary_loss_mlp": 0.01029653,
      "balance_loss_clip": 1.05144119,
      "balance_loss_mlp": 1.02135599,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 2.6638792274545215,
      "language_loss": 0.82047057,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.84240174,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.675306558609009
    },
    {
      "auxiliary_loss_clip": 0.01148929,
      "auxiliary_loss_mlp": 0.01023244,
      "balance_loss_clip": 1.04328001,
      "balance_loss_mlp": 1.01557279,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 2.5357616296615175,
      "language_loss": 0.76780725,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78952897,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.747596263885498
    },
    {
      "auxiliary_loss_clip": 0.01157627,
      "auxiliary_loss_mlp": 0.01030146,
      "balance_loss_clip": 1.04735899,
      "balance_loss_mlp": 1.02271295,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.327721445676804,
      "language_loss": 0.68028498,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70216274,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.8013575077056885
    },
    {
      "auxiliary_loss_clip": 0.01151673,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.04471779,
      "balance_loss_mlp": 1.02061832,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.229174550310233,
      "language_loss": 0.66704178,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68884587,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.829495429992676
    },
    {
      "auxiliary_loss_clip": 0.01159929,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.04670238,
      "balance_loss_mlp": 1.02160645,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.305951271063803,
      "language_loss": 0.75418389,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77606827,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 3.5647542476654053
    },
    {
      "auxiliary_loss_clip": 0.0115241,
      "auxiliary_loss_mlp": 0.01025176,
      "balance_loss_clip": 1.04865956,
      "balance_loss_mlp": 1.01799977,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 2.2232754179174323,
      "language_loss": 0.74273539,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76451123,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.7235939502716064
    },
    {
      "auxiliary_loss_clip": 0.01166521,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 1.05081308,
      "balance_loss_mlp": 1.019063,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.9302591755074887,
      "language_loss": 0.76772034,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78965062,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.628178119659424
    },
    {
      "auxiliary_loss_clip": 0.01172134,
      "auxiliary_loss_mlp": 0.01024734,
      "balance_loss_clip": 1.05035186,
      "balance_loss_mlp": 1.01708102,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 2.3436049958426044,
      "language_loss": 0.78851694,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.8104856,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 2.6766562461853027
    },
    {
      "auxiliary_loss_clip": 0.01176924,
      "auxiliary_loss_mlp": 0.01030559,
      "balance_loss_clip": 1.05073833,
      "balance_loss_mlp": 1.02244639,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 2.6050122399930715,
      "language_loss": 0.80133712,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.82341194,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.572788953781128
    },
    {
      "auxiliary_loss_clip": 0.01173931,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.05011082,
      "balance_loss_mlp": 1.01928258,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 2.270460712389649,
      "language_loss": 0.57127905,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59328228,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.718236207962036
    },
    {
      "auxiliary_loss_clip": 0.01140044,
      "auxiliary_loss_mlp": 0.01029139,
      "balance_loss_clip": 1.04772949,
      "balance_loss_mlp": 1.02106845,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.231593653633515,
      "language_loss": 0.77641129,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79810321,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 2.6872613430023193
    },
    {
      "auxiliary_loss_clip": 0.01177473,
      "auxiliary_loss_mlp": 0.01024369,
      "balance_loss_clip": 1.04998779,
      "balance_loss_mlp": 1.01625121,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 2.572809399092045,
      "language_loss": 0.79343724,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81545568,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.6353001594543457
    },
    {
      "auxiliary_loss_clip": 0.01168965,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.0482986,
      "balance_loss_mlp": 1.02105355,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.6761709811725838,
      "language_loss": 0.84293985,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86491883,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 3.5795416831970215
    },
    {
      "auxiliary_loss_clip": 0.01149147,
      "auxiliary_loss_mlp": 0.01021224,
      "balance_loss_clip": 1.04252887,
      "balance_loss_mlp": 1.01307058,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 2.0677431767322263,
      "language_loss": 0.73978198,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.7614857,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.930359125137329
    },
    {
      "auxiliary_loss_clip": 0.01149204,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.04556894,
      "balance_loss_mlp": 1.02064919,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 2.9048344246278535,
      "language_loss": 0.7053138,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.72709513,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.769329786300659
    },
    {
      "auxiliary_loss_clip": 0.01159415,
      "auxiliary_loss_mlp": 0.01025338,
      "balance_loss_clip": 1.04896915,
      "balance_loss_mlp": 1.01774466,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.8104942550464072,
      "language_loss": 0.75920302,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78105056,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.692862033843994
    },
    {
      "auxiliary_loss_clip": 0.01154595,
      "auxiliary_loss_mlp": 0.01028365,
      "balance_loss_clip": 1.04790306,
      "balance_loss_mlp": 1.02059221,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 5.900929220336653,
      "language_loss": 0.77702343,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.79885304,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 3.488489866256714
    },
    {
      "auxiliary_loss_clip": 0.01155258,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.04948711,
      "balance_loss_mlp": 1.02064347,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 1.9775158750274546,
      "language_loss": 0.70195222,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72378576,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.743899345397949
    },
    {
      "auxiliary_loss_clip": 0.01155513,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.05066013,
      "balance_loss_mlp": 1.0171349,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.7470444974093478,
      "language_loss": 0.82206917,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84387153,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.663330554962158
    },
    {
      "auxiliary_loss_clip": 0.01134609,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.04515398,
      "balance_loss_mlp": 1.01532602,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.133514959129627,
      "language_loss": 0.80039096,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82196808,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.679927110671997
    },
    {
      "auxiliary_loss_clip": 0.01157018,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.04683948,
      "balance_loss_mlp": 1.01998568,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.526553076191352,
      "language_loss": 0.76427865,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.7861256,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.6686761379241943
    },
    {
      "auxiliary_loss_clip": 0.01161806,
      "auxiliary_loss_mlp": 0.01024013,
      "balance_loss_clip": 1.0486691,
      "balance_loss_mlp": 1.01653278,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 1.7957564567979554,
      "language_loss": 0.77277035,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79462856,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 3.573181629180908
    },
    {
      "auxiliary_loss_clip": 0.01167271,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.05119133,
      "balance_loss_mlp": 1.02004886,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.3920269608943325,
      "language_loss": 0.86051238,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88246149,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.6084189414978027
    },
    {
      "auxiliary_loss_clip": 0.01125543,
      "auxiliary_loss_mlp": 0.01022381,
      "balance_loss_clip": 1.041556,
      "balance_loss_mlp": 1.0142417,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.8253969880781373,
      "language_loss": 0.78952026,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.81099951,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.7743723392486572
    },
    {
      "auxiliary_loss_clip": 0.01144032,
      "auxiliary_loss_mlp": 0.01021118,
      "balance_loss_clip": 1.04670286,
      "balance_loss_mlp": 1.01378679,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.8836766435944754,
      "language_loss": 0.74986577,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77151728,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.7517752647399902
    },
    {
      "auxiliary_loss_clip": 0.01177486,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.05381262,
      "balance_loss_mlp": 1.02038991,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.3170843745261758,
      "language_loss": 0.70236462,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72441649,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.6806085109710693
    },
    {
      "auxiliary_loss_clip": 0.01164107,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.05074656,
      "balance_loss_mlp": 1.01908016,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.3644037087799306,
      "language_loss": 0.76845795,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.79036283,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.634230852127075
    },
    {
      "auxiliary_loss_clip": 0.01178561,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.05037796,
      "balance_loss_mlp": 1.02201807,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.3462149585813816,
      "language_loss": 0.80628556,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82837182,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.6034460067749023
    },
    {
      "auxiliary_loss_clip": 0.01076149,
      "auxiliary_loss_mlp": 0.01003113,
      "balance_loss_clip": 1.01697195,
      "balance_loss_mlp": 1.00215387,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8143195208259681,
      "language_loss": 0.58208704,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.6028797,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.3281984329223633
    },
    {
      "auxiliary_loss_clip": 0.011457,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.04424,
      "balance_loss_mlp": 1.02090776,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 2.336565907321993,
      "language_loss": 0.80187273,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.8236208,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.76492977142334
    },
    {
      "auxiliary_loss_clip": 0.01164894,
      "auxiliary_loss_mlp": 0.01022639,
      "balance_loss_clip": 1.04960537,
      "balance_loss_mlp": 1.01558149,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.700598636342484,
      "language_loss": 0.92024922,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.94212455,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 3.5921247005462646
    },
    {
      "auxiliary_loss_clip": 0.01164636,
      "auxiliary_loss_mlp": 0.00901281,
      "balance_loss_clip": 1.04794824,
      "balance_loss_mlp": 1.00063515,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 4.124467658287821,
      "language_loss": 0.84456825,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86522746,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.7207372188568115
    },
    {
      "auxiliary_loss_clip": 0.01175499,
      "auxiliary_loss_mlp": 0.00901761,
      "balance_loss_clip": 1.04941559,
      "balance_loss_mlp": 1.00072396,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 2.3799726674431576,
      "language_loss": 0.68398297,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.70475554,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.5930914878845215
    },
    {
      "auxiliary_loss_clip": 0.0116955,
      "auxiliary_loss_mlp": 0.01025257,
      "balance_loss_clip": 1.04834414,
      "balance_loss_mlp": 1.0176369,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.7130999391783346,
      "language_loss": 0.70177448,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72372246,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.6169657707214355
    },
    {
      "auxiliary_loss_clip": 0.01166144,
      "auxiliary_loss_mlp": 0.0090103,
      "balance_loss_clip": 1.05079472,
      "balance_loss_mlp": 1.00063086,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.108480466635864,
      "language_loss": 0.79794776,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.81861949,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 2.648483991622925
    },
    {
      "auxiliary_loss_clip": 0.01150944,
      "auxiliary_loss_mlp": 0.01029594,
      "balance_loss_clip": 1.04541326,
      "balance_loss_mlp": 1.02164888,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 4.623964874725401,
      "language_loss": 0.71198952,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.73379493,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.696876049041748
    },
    {
      "auxiliary_loss_clip": 0.01156073,
      "auxiliary_loss_mlp": 0.01025875,
      "balance_loss_clip": 1.04837871,
      "balance_loss_mlp": 1.01873457,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 2.3734603102358913,
      "language_loss": 0.78755522,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80937469,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.7592661380767822
    },
    {
      "auxiliary_loss_clip": 0.01165323,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 1.0496465,
      "balance_loss_mlp": 1.01638079,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 5.2258025160916075,
      "language_loss": 0.78090358,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80279678,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.654944658279419
    },
    {
      "auxiliary_loss_clip": 0.01165142,
      "auxiliary_loss_mlp": 0.01024126,
      "balance_loss_clip": 1.04919386,
      "balance_loss_mlp": 1.01711965,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 2.3161306225771727,
      "language_loss": 0.83053035,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.85242301,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.569809913635254
    },
    {
      "auxiliary_loss_clip": 0.01157254,
      "auxiliary_loss_mlp": 0.00901981,
      "balance_loss_clip": 1.04955816,
      "balance_loss_mlp": 1.00084829,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.8424455589695756,
      "language_loss": 0.78099346,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80158591,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 3.6426446437835693
    },
    {
      "auxiliary_loss_clip": 0.01138045,
      "auxiliary_loss_mlp": 0.01028023,
      "balance_loss_clip": 1.04479313,
      "balance_loss_mlp": 1.02019668,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.586327047387064,
      "language_loss": 0.56603205,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.58769274,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.75875186920166
    },
    {
      "auxiliary_loss_clip": 0.01166527,
      "auxiliary_loss_mlp": 0.0102682,
      "balance_loss_clip": 1.05299497,
      "balance_loss_mlp": 1.0187912,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 2.026827742482432,
      "language_loss": 0.74286908,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76480252,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.6548879146575928
    },
    {
      "auxiliary_loss_clip": 0.01140345,
      "auxiliary_loss_mlp": 0.01032049,
      "balance_loss_clip": 1.04362512,
      "balance_loss_mlp": 1.0240202,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 1.8712245314998455,
      "language_loss": 0.74527854,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76700246,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 3.7655608654022217
    },
    {
      "auxiliary_loss_clip": 0.01164048,
      "auxiliary_loss_mlp": 0.01029936,
      "balance_loss_clip": 1.05327487,
      "balance_loss_mlp": 1.02143657,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.6930342760764636,
      "language_loss": 0.68814874,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71008861,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.819127082824707
    },
    {
      "auxiliary_loss_clip": 0.01176123,
      "auxiliary_loss_mlp": 0.01021551,
      "balance_loss_clip": 1.05054927,
      "balance_loss_mlp": 1.01399314,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 2.8230504138888395,
      "language_loss": 0.74698532,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76896203,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.6291439533233643
    },
    {
      "auxiliary_loss_clip": 0.01066671,
      "auxiliary_loss_mlp": 0.01000863,
      "balance_loss_clip": 1.0146786,
      "balance_loss_mlp": 0.9998619,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.773437654601755,
      "language_loss": 0.55709261,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57776797,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.257993459701538
    },
    {
      "auxiliary_loss_clip": 0.01159959,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.04748762,
      "balance_loss_mlp": 1.02113783,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.9140815395291135,
      "language_loss": 0.68174422,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70362866,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.6010334491729736
    },
    {
      "auxiliary_loss_clip": 0.01177683,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.05006289,
      "balance_loss_mlp": 1.02060509,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 2.139747065841366,
      "language_loss": 0.81290638,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83497316,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 3.546358823776245
    },
    {
      "auxiliary_loss_clip": 0.0114643,
      "auxiliary_loss_mlp": 0.01027285,
      "balance_loss_clip": 1.04591131,
      "balance_loss_mlp": 1.02024007,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 2.20449811115197,
      "language_loss": 0.60162914,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.6233663,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.8152055740356445
    },
    {
      "auxiliary_loss_clip": 0.01143924,
      "auxiliary_loss_mlp": 0.01024624,
      "balance_loss_clip": 1.04365599,
      "balance_loss_mlp": 1.01706576,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.9379925795769013,
      "language_loss": 0.64497274,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.66665822,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.7742817401885986
    },
    {
      "auxiliary_loss_clip": 0.01156956,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.04889762,
      "balance_loss_mlp": 1.02169275,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 6.46735560007225,
      "language_loss": 0.77691519,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.79877281,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.6227829456329346
    },
    {
      "auxiliary_loss_clip": 0.01135457,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.04124916,
      "balance_loss_mlp": 1.02041733,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 1.778113750785284,
      "language_loss": 0.77665424,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79828823,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.7635562419891357
    },
    {
      "auxiliary_loss_clip": 0.01142448,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.04511976,
      "balance_loss_mlp": 1.01869845,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.0445317061496353,
      "language_loss": 0.79293072,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81462622,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.6567654609680176
    },
    {
      "auxiliary_loss_clip": 0.01162471,
      "auxiliary_loss_mlp": 0.01030176,
      "balance_loss_clip": 1.04723358,
      "balance_loss_mlp": 1.02281141,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.995483420280117,
      "language_loss": 0.68215454,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.704081,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.6929845809936523
    },
    {
      "auxiliary_loss_clip": 0.01174042,
      "auxiliary_loss_mlp": 0.01024635,
      "balance_loss_clip": 1.05165291,
      "balance_loss_mlp": 1.01736891,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 2.6071908752782345,
      "language_loss": 0.82542616,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84741294,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.565896511077881
    },
    {
      "auxiliary_loss_clip": 0.01133146,
      "auxiliary_loss_mlp": 0.01027077,
      "balance_loss_clip": 1.04537952,
      "balance_loss_mlp": 1.01998162,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.261333037743372,
      "language_loss": 0.78722697,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.80882919,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.7633168697357178
    },
    {
      "auxiliary_loss_clip": 0.01172054,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 1.04804134,
      "balance_loss_mlp": 1.01558769,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.6754783234980373,
      "language_loss": 0.81339419,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83534783,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.6405584812164307
    },
    {
      "auxiliary_loss_clip": 0.01139963,
      "auxiliary_loss_mlp": 0.01023905,
      "balance_loss_clip": 1.04689217,
      "balance_loss_mlp": 1.01548886,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.824803699152641,
      "language_loss": 0.69173145,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.71337008,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 4.070333480834961
    },
    {
      "auxiliary_loss_clip": 0.01166622,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.0516628,
      "balance_loss_mlp": 1.0162499,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 2.006743003122044,
      "language_loss": 0.80500352,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.82690436,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.6669280529022217
    },
    {
      "auxiliary_loss_clip": 0.01144144,
      "auxiliary_loss_mlp": 0.0102528,
      "balance_loss_clip": 1.04550683,
      "balance_loss_mlp": 1.01744819,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.316474061970433,
      "language_loss": 0.82750136,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84919554,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.683485269546509
    },
    {
      "auxiliary_loss_clip": 0.01159716,
      "auxiliary_loss_mlp": 0.01022301,
      "balance_loss_clip": 1.04964161,
      "balance_loss_mlp": 1.01521969,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.9605823745242632,
      "language_loss": 0.78891897,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.81073916,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.615469455718994
    },
    {
      "auxiliary_loss_clip": 0.01155805,
      "auxiliary_loss_mlp": 0.01025618,
      "balance_loss_clip": 1.04647446,
      "balance_loss_mlp": 1.01771474,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.6579013485521432,
      "language_loss": 0.79078406,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81259829,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.7575321197509766
    },
    {
      "auxiliary_loss_clip": 0.01129892,
      "auxiliary_loss_mlp": 0.01023723,
      "balance_loss_clip": 1.04084492,
      "balance_loss_mlp": 1.01570606,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.6303180088953702,
      "language_loss": 0.81167603,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.8332122,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.6949405670166016
    },
    {
      "auxiliary_loss_clip": 0.01168185,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.04843163,
      "balance_loss_mlp": 1.02457082,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.8379190444779314,
      "language_loss": 0.83720326,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.85921317,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.751146078109741
    },
    {
      "auxiliary_loss_clip": 0.01156546,
      "auxiliary_loss_mlp": 0.01025081,
      "balance_loss_clip": 1.04709303,
      "balance_loss_mlp": 1.01719534,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.2923630659953287,
      "language_loss": 0.75476968,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77658594,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.681138277053833
    },
    {
      "auxiliary_loss_clip": 0.01159354,
      "auxiliary_loss_mlp": 0.0102313,
      "balance_loss_clip": 1.04688299,
      "balance_loss_mlp": 1.01577771,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.9725142654907788,
      "language_loss": 0.75026357,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77208841,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 3.5483648777008057
    },
    {
      "auxiliary_loss_clip": 0.01166667,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.04756856,
      "balance_loss_mlp": 1.01896203,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 2.1197459119379776,
      "language_loss": 0.67961323,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70155084,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.6097488403320312
    },
    {
      "auxiliary_loss_clip": 0.01168542,
      "auxiliary_loss_mlp": 0.01027406,
      "balance_loss_clip": 1.05073214,
      "balance_loss_mlp": 1.01971674,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 2.026135267121629,
      "language_loss": 0.84779489,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.86975437,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.6376607418060303
    },
    {
      "auxiliary_loss_clip": 0.01168607,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.04931104,
      "balance_loss_mlp": 1.01947832,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 1.9581021527381741,
      "language_loss": 0.82769334,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84965527,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.5756990909576416
    },
    {
      "auxiliary_loss_clip": 0.01148408,
      "auxiliary_loss_mlp": 0.01023687,
      "balance_loss_clip": 1.04549503,
      "balance_loss_mlp": 1.01642132,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 1.8341977951365485,
      "language_loss": 0.84036267,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86208367,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 3.71693754196167
    },
    {
      "auxiliary_loss_clip": 0.01067868,
      "auxiliary_loss_mlp": 0.0100069,
      "balance_loss_clip": 1.01578474,
      "balance_loss_mlp": 0.99969465,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7083747221016075,
      "language_loss": 0.54226953,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56295514,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.277463674545288
    },
    {
      "auxiliary_loss_clip": 0.01156448,
      "auxiliary_loss_mlp": 0.00901956,
      "balance_loss_clip": 1.04692483,
      "balance_loss_mlp": 1.00077748,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.2422153216011376,
      "language_loss": 0.78442568,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80500972,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.7177703380584717
    },
    {
      "auxiliary_loss_clip": 0.01155746,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04725015,
      "balance_loss_mlp": 1.01758206,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.4773198069558666,
      "language_loss": 0.70675933,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.72856885,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.717048406600952
    },
    {
      "auxiliary_loss_clip": 0.01171378,
      "auxiliary_loss_mlp": 0.01021846,
      "balance_loss_clip": 1.05022466,
      "balance_loss_mlp": 1.0149138,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.7894149382908955,
      "language_loss": 0.62970424,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.65163648,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.662477731704712
    },
    {
      "auxiliary_loss_clip": 0.0113772,
      "auxiliary_loss_mlp": 0.0102921,
      "balance_loss_clip": 1.04123795,
      "balance_loss_mlp": 1.02121079,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.3510931407025413,
      "language_loss": 0.74098951,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76265877,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 3.7010419368743896
    },
    {
      "auxiliary_loss_clip": 0.0116764,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 1.05144966,
      "balance_loss_mlp": 1.0205977,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.9775169915279007,
      "language_loss": 0.73172003,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75367713,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.583237409591675
    },
    {
      "auxiliary_loss_clip": 0.01148027,
      "auxiliary_loss_mlp": 0.0103104,
      "balance_loss_clip": 1.04750681,
      "balance_loss_mlp": 1.02286863,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 1.81640222500929,
      "language_loss": 0.62764502,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.64943564,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.7221274375915527
    },
    {
      "auxiliary_loss_clip": 0.01154771,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.04649115,
      "balance_loss_mlp": 1.01569557,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 1.8131974780486748,
      "language_loss": 0.7240178,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.7457962,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.6843338012695312
    },
    {
      "auxiliary_loss_clip": 0.01176978,
      "auxiliary_loss_mlp": 0.01025879,
      "balance_loss_clip": 1.05149508,
      "balance_loss_mlp": 1.01848817,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.684866730495991,
      "language_loss": 0.72658092,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74860948,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.6467010974884033
    },
    {
      "auxiliary_loss_clip": 0.01149569,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.04792011,
      "balance_loss_mlp": 1.0181303,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.7560911999949222,
      "language_loss": 0.75816983,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77992344,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.7321078777313232
    },
    {
      "auxiliary_loss_clip": 0.01176318,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 1.05216944,
      "balance_loss_mlp": 1.01929855,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 2.5471077957492603,
      "language_loss": 0.73652136,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75854862,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.614062547683716
    },
    {
      "auxiliary_loss_clip": 0.01161484,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04809356,
      "balance_loss_mlp": 1.01723552,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.629694226104805,
      "language_loss": 0.71163225,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73349476,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.6473052501678467
    },
    {
      "auxiliary_loss_clip": 0.01150901,
      "auxiliary_loss_mlp": 0.01023416,
      "balance_loss_clip": 1.04548883,
      "balance_loss_mlp": 1.01602483,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.950357967313402,
      "language_loss": 0.78449869,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80624193,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.6997387409210205
    },
    {
      "auxiliary_loss_clip": 0.01160511,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.0472877,
      "balance_loss_mlp": 1.0209403,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.6440704974326543,
      "language_loss": 0.82068866,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.8425802,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 3.6021368503570557
    },
    {
      "auxiliary_loss_clip": 0.01079314,
      "auxiliary_loss_mlp": 0.01000977,
      "balance_loss_clip": 1.01574385,
      "balance_loss_mlp": 1.00006521,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7658448037554233,
      "language_loss": 0.5685122,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58931506,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.1972429752349854
    },
    {
      "auxiliary_loss_clip": 0.0112366,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.04252219,
      "balance_loss_mlp": 1.02123582,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.9114184999379091,
      "language_loss": 0.71384323,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73537433,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 2.676640272140503
    },
    {
      "auxiliary_loss_clip": 0.01150796,
      "auxiliary_loss_mlp": 0.0102407,
      "balance_loss_clip": 1.04723859,
      "balance_loss_mlp": 1.01702785,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.6812342590503249,
      "language_loss": 0.6360116,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65776026,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.6907882690429688
    },
    {
      "auxiliary_loss_clip": 0.0115541,
      "auxiliary_loss_mlp": 0.01022712,
      "balance_loss_clip": 1.04185796,
      "balance_loss_mlp": 1.01489186,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.796307493900582,
      "language_loss": 0.69015169,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.7119329,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.652327060699463
    },
    {
      "auxiliary_loss_clip": 0.0117771,
      "auxiliary_loss_mlp": 0.01022191,
      "balance_loss_clip": 1.05228317,
      "balance_loss_mlp": 1.01410866,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.0510762000536036,
      "language_loss": 0.75821674,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.78021574,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.639266014099121
    },
    {
      "auxiliary_loss_clip": 0.01149666,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.04996502,
      "balance_loss_mlp": 1.02064741,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.273276408603987,
      "language_loss": 0.77470493,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79648936,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.6787185668945312
    },
    {
      "auxiliary_loss_clip": 0.01150839,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.04332995,
      "balance_loss_mlp": 1.01630592,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 2.084045856289613,
      "language_loss": 0.83359861,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85534942,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 2.6342809200286865
    },
    {
      "auxiliary_loss_clip": 0.01174316,
      "auxiliary_loss_mlp": 0.01025051,
      "balance_loss_clip": 1.05064118,
      "balance_loss_mlp": 1.01749921,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 1.819794205507839,
      "language_loss": 0.68482149,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70681512,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.497737407684326
    },
    {
      "auxiliary_loss_clip": 0.01170215,
      "auxiliary_loss_mlp": 0.01020676,
      "balance_loss_clip": 1.04714394,
      "balance_loss_mlp": 1.01256371,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.110439831148686,
      "language_loss": 0.77359831,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79550719,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 3.6309642791748047
    },
    {
      "auxiliary_loss_clip": 0.01143447,
      "auxiliary_loss_mlp": 0.01023866,
      "balance_loss_clip": 1.04489052,
      "balance_loss_mlp": 1.01605177,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 1.9894452681987456,
      "language_loss": 0.64791727,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.66959041,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.6807544231414795
    },
    {
      "auxiliary_loss_clip": 0.01158877,
      "auxiliary_loss_mlp": 0.01024532,
      "balance_loss_clip": 1.04622984,
      "balance_loss_mlp": 1.01714683,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 2.0758653344279003,
      "language_loss": 0.77413535,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79596943,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.6724913120269775
    },
    {
      "auxiliary_loss_clip": 0.01150626,
      "auxiliary_loss_mlp": 0.00901733,
      "balance_loss_clip": 1.04582977,
      "balance_loss_mlp": 1.00079417,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 1.8373010077457983,
      "language_loss": 0.7954694,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81599295,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.6907358169555664
    },
    {
      "auxiliary_loss_clip": 0.01166413,
      "auxiliary_loss_mlp": 0.01024876,
      "balance_loss_clip": 1.04981279,
      "balance_loss_mlp": 1.01718068,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 1.7985787792684014,
      "language_loss": 0.62034678,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64225972,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 3.6826138496398926
    },
    {
      "auxiliary_loss_clip": 0.01137511,
      "auxiliary_loss_mlp": 0.01023858,
      "balance_loss_clip": 1.0445528,
      "balance_loss_mlp": 1.01609159,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.4420440962938985,
      "language_loss": 0.68019277,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.70180643,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.6642608642578125
    },
    {
      "auxiliary_loss_clip": 0.01148828,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.0469501,
      "balance_loss_mlp": 1.01886201,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.644458956322948,
      "language_loss": 0.77954924,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80130315,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.7267298698425293
    },
    {
      "auxiliary_loss_clip": 0.01168901,
      "auxiliary_loss_mlp": 0.01034746,
      "balance_loss_clip": 1.05321097,
      "balance_loss_mlp": 1.02618051,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.252602059161334,
      "language_loss": 0.77880824,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.80084467,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.613147735595703
    },
    {
      "auxiliary_loss_clip": 0.01160143,
      "auxiliary_loss_mlp": 0.01023471,
      "balance_loss_clip": 1.04481435,
      "balance_loss_mlp": 1.0155375,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 2.675738770001643,
      "language_loss": 0.49405226,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.51588845,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.6401257514953613
    },
    {
      "auxiliary_loss_clip": 0.01171215,
      "auxiliary_loss_mlp": 0.01027347,
      "balance_loss_clip": 1.04762697,
      "balance_loss_mlp": 1.01962864,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.0938336219151052,
      "language_loss": 0.82663214,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.84861767,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 3.49271821975708
    },
    {
      "auxiliary_loss_clip": 0.01079285,
      "auxiliary_loss_mlp": 0.01000786,
      "balance_loss_clip": 1.01506019,
      "balance_loss_mlp": 0.99977916,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7575970035898811,
      "language_loss": 0.55041075,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57121146,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.0189225673675537
    },
    {
      "auxiliary_loss_clip": 0.01141617,
      "auxiliary_loss_mlp": 0.01028606,
      "balance_loss_clip": 1.04435253,
      "balance_loss_mlp": 1.02125382,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.7740840406927576,
      "language_loss": 0.68106222,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70276439,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.6920549869537354
    },
    {
      "auxiliary_loss_clip": 0.01155106,
      "auxiliary_loss_mlp": 0.0102149,
      "balance_loss_clip": 1.04583073,
      "balance_loss_mlp": 1.01459944,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.0399921610702547,
      "language_loss": 0.71339142,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.73515749,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.680835485458374
    },
    {
      "auxiliary_loss_clip": 0.01160838,
      "auxiliary_loss_mlp": 0.01030263,
      "balance_loss_clip": 1.0464685,
      "balance_loss_mlp": 1.02210855,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 2.6198985029257864,
      "language_loss": 0.7558285,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77773958,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.7372984886169434
    },
    {
      "auxiliary_loss_clip": 0.01074529,
      "auxiliary_loss_mlp": 0.01000154,
      "balance_loss_clip": 1.01430929,
      "balance_loss_mlp": 0.99910539,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7898819729753951,
      "language_loss": 0.59302366,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61377048,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.0268032550811768
    },
    {
      "auxiliary_loss_clip": 0.01165228,
      "auxiliary_loss_mlp": 0.01023518,
      "balance_loss_clip": 1.04985261,
      "balance_loss_mlp": 1.01583481,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.0116761148562077,
      "language_loss": 0.8091414,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.83102882,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.600257396697998
    },
    {
      "auxiliary_loss_clip": 0.01165008,
      "auxiliary_loss_mlp": 0.01031248,
      "balance_loss_clip": 1.04918838,
      "balance_loss_mlp": 1.02373505,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.758479620245624,
      "language_loss": 0.77477753,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79674011,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.599959135055542
    },
    {
      "auxiliary_loss_clip": 0.01156967,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04402149,
      "balance_loss_mlp": 1.01676869,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.655196003190879,
      "language_loss": 0.92153138,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94334996,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 3.6534624099731445
    },
    {
      "auxiliary_loss_clip": 0.01138772,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.0445838,
      "balance_loss_mlp": 1.02319741,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.2507463176708904,
      "language_loss": 0.8676281,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.88933188,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.699605703353882
    },
    {
      "auxiliary_loss_clip": 0.01137977,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.04300153,
      "balance_loss_mlp": 1.01911914,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.712870765261427,
      "language_loss": 0.80201364,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82366049,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.7023720741271973
    },
    {
      "auxiliary_loss_clip": 0.01149689,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.04498434,
      "balance_loss_mlp": 1.02093494,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 1.805162136743757,
      "language_loss": 0.77428138,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79606014,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.8137784004211426
    },
    {
      "auxiliary_loss_clip": 0.011525,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 1.04585361,
      "balance_loss_mlp": 1.01855159,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.1227418478569855,
      "language_loss": 0.77235842,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79414469,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.6580569744110107
    },
    {
      "auxiliary_loss_clip": 0.01143103,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04367757,
      "balance_loss_mlp": 1.01698041,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 1.9896884583738945,
      "language_loss": 0.75723886,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.77891457,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.7375640869140625
    },
    {
      "auxiliary_loss_clip": 0.01144643,
      "auxiliary_loss_mlp": 0.01027996,
      "balance_loss_clip": 1.04234672,
      "balance_loss_mlp": 1.01951993,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.6192247880990918,
      "language_loss": 0.78407663,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.805803,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.6641645431518555
    },
    {
      "auxiliary_loss_clip": 0.01152243,
      "auxiliary_loss_mlp": 0.01030175,
      "balance_loss_clip": 1.0473547,
      "balance_loss_mlp": 1.02249813,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 2.1031190563395312,
      "language_loss": 0.79840791,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.82023215,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6678965091705322
    },
    {
      "auxiliary_loss_clip": 0.01135621,
      "auxiliary_loss_mlp": 0.01028426,
      "balance_loss_clip": 1.04341984,
      "balance_loss_mlp": 1.02085006,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.997443739110924,
      "language_loss": 0.74682081,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.76846129,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.6932525634765625
    },
    {
      "auxiliary_loss_clip": 0.01154449,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.05028582,
      "balance_loss_mlp": 1.02180576,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 3.713986767173612,
      "language_loss": 0.81658125,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.8384198,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 3.5799126625061035
    },
    {
      "auxiliary_loss_clip": 0.01162333,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.04862344,
      "balance_loss_mlp": 1.02036774,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 2.240425882663679,
      "language_loss": 0.89049971,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.91240227,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.5295493602752686
    },
    {
      "auxiliary_loss_clip": 0.01071545,
      "auxiliary_loss_mlp": 0.01003818,
      "balance_loss_clip": 1.01377153,
      "balance_loss_mlp": 1.00284088,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7588436334363077,
      "language_loss": 0.6074546,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62820822,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 2.9424052238464355
    },
    {
      "auxiliary_loss_clip": 0.01149933,
      "auxiliary_loss_mlp": 0.01026257,
      "balance_loss_clip": 1.04254007,
      "balance_loss_mlp": 1.01813245,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.2249586122478284,
      "language_loss": 0.78775024,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.80951214,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.585583209991455
    },
    {
      "auxiliary_loss_clip": 0.01167473,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.04970813,
      "balance_loss_mlp": 1.02009916,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.294562232207061,
      "language_loss": 0.6458571,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.66781092,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 3.3060617446899414
    },
    {
      "auxiliary_loss_clip": 0.0107125,
      "auxiliary_loss_mlp": 0.01001954,
      "balance_loss_clip": 1.0133338,
      "balance_loss_mlp": 1.00101244,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7524438954950575,
      "language_loss": 0.51820344,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53893548,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.0663986206054688
    },
    {
      "auxiliary_loss_clip": 0.01148939,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04344559,
      "balance_loss_mlp": 1.01701975,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 4.990517159605142,
      "language_loss": 0.69029927,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.71204054,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.6366047859191895
    },
    {
      "auxiliary_loss_clip": 0.0116449,
      "auxiliary_loss_mlp": 0.0103139,
      "balance_loss_clip": 1.05347586,
      "balance_loss_mlp": 1.02282476,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 4.611743596037231,
      "language_loss": 0.74647784,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76843667,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.6908481121063232
    },
    {
      "auxiliary_loss_clip": 0.01162226,
      "auxiliary_loss_mlp": 0.0101992,
      "balance_loss_clip": 1.05143476,
      "balance_loss_mlp": 1.0123682,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.9009366439462574,
      "language_loss": 0.85659051,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87841195,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.6627774238586426
    },
    {
      "auxiliary_loss_clip": 0.01142094,
      "auxiliary_loss_mlp": 0.00901241,
      "balance_loss_clip": 1.04657543,
      "balance_loss_mlp": 1.00070941,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 21.671196373547634,
      "language_loss": 0.70687628,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.7273097,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 3.793602228164673
    },
    {
      "auxiliary_loss_clip": 0.01166036,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.05018926,
      "balance_loss_mlp": 1.01677668,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 2.014546454748834,
      "language_loss": 0.72043788,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.74234414,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.624774217605591
    },
    {
      "auxiliary_loss_clip": 0.01106853,
      "auxiliary_loss_mlp": 0.01022859,
      "balance_loss_clip": 1.03808999,
      "balance_loss_mlp": 1.01572454,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.1827829689829104,
      "language_loss": 0.83857691,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85987401,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.7442476749420166
    },
    {
      "auxiliary_loss_clip": 0.01142423,
      "auxiliary_loss_mlp": 0.01024896,
      "balance_loss_clip": 1.04536712,
      "balance_loss_mlp": 1.01708198,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.8737952993272429,
      "language_loss": 0.78676462,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80843782,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.782158136367798
    },
    {
      "auxiliary_loss_clip": 0.01177442,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.04974842,
      "balance_loss_mlp": 1.02537012,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.0741615639482984,
      "language_loss": 0.74739838,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76950717,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.5321249961853027
    },
    {
      "auxiliary_loss_clip": 0.01156695,
      "auxiliary_loss_mlp": 0.01026768,
      "balance_loss_clip": 1.04594624,
      "balance_loss_mlp": 1.01940691,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.3836622666361884,
      "language_loss": 0.78455365,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80638826,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.5801870822906494
    },
    {
      "auxiliary_loss_clip": 0.01173953,
      "auxiliary_loss_mlp": 0.01027606,
      "balance_loss_clip": 1.05149198,
      "balance_loss_mlp": 1.02030158,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 2.4488277573452004,
      "language_loss": 0.80384296,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82585859,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.5947725772857666
    },
    {
      "auxiliary_loss_clip": 0.01158694,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04495001,
      "balance_loss_mlp": 1.01865852,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.239341412374733,
      "language_loss": 0.71401656,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.73586571,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.6399521827697754
    },
    {
      "auxiliary_loss_clip": 0.01157427,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.04558086,
      "balance_loss_mlp": 1.02058661,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 2.4362258837642763,
      "language_loss": 0.66575706,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68761915,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.6206393241882324
    },
    {
      "auxiliary_loss_clip": 0.01175316,
      "auxiliary_loss_mlp": 0.00901807,
      "balance_loss_clip": 1.05081451,
      "balance_loss_mlp": 1.00075841,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 2.470727575024006,
      "language_loss": 0.62807965,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.64885092,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 3.5620908737182617
    },
    {
      "auxiliary_loss_clip": 0.01176892,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 1.05196452,
      "balance_loss_mlp": 1.0194943,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.17368158490422,
      "language_loss": 0.78668845,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80872858,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.5386767387390137
    },
    {
      "auxiliary_loss_clip": 0.01162491,
      "auxiliary_loss_mlp": 0.01031238,
      "balance_loss_clip": 1.04663098,
      "balance_loss_mlp": 1.02426457,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.6543770045693975,
      "language_loss": 0.69259059,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71452785,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.5886075496673584
    },
    {
      "auxiliary_loss_clip": 0.01155254,
      "auxiliary_loss_mlp": 0.01031865,
      "balance_loss_clip": 1.05025351,
      "balance_loss_mlp": 1.02344322,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 2.1026287711516813,
      "language_loss": 0.70774698,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72961819,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.6830015182495117
    },
    {
      "auxiliary_loss_clip": 0.01170339,
      "auxiliary_loss_mlp": 0.01029584,
      "balance_loss_clip": 1.04916787,
      "balance_loss_mlp": 1.02101278,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 2.010360223693067,
      "language_loss": 0.82383639,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84583569,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.6405932903289795
    },
    {
      "auxiliary_loss_clip": 0.01158653,
      "auxiliary_loss_mlp": 0.01030133,
      "balance_loss_clip": 1.04780877,
      "balance_loss_mlp": 1.02270651,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 3.2632245946922445,
      "language_loss": 0.70674706,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72863489,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.6799662113189697
    },
    {
      "auxiliary_loss_clip": 0.01157439,
      "auxiliary_loss_mlp": 0.01026887,
      "balance_loss_clip": 1.04680836,
      "balance_loss_mlp": 1.01885223,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 2.764482984464052,
      "language_loss": 0.61294186,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63478518,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.5964033603668213
    },
    {
      "auxiliary_loss_clip": 0.01052583,
      "auxiliary_loss_mlp": 0.0100165,
      "balance_loss_clip": 1.0139699,
      "balance_loss_mlp": 1.00070214,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 2.5623185839187808,
      "language_loss": 0.62502354,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64556587,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.2196624279022217
    },
    {
      "auxiliary_loss_clip": 0.01176023,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.0510596,
      "balance_loss_mlp": 1.02018094,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 2.3941150941018785,
      "language_loss": 0.74691439,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76895165,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 3.5260486602783203
    },
    {
      "auxiliary_loss_clip": 0.01146374,
      "auxiliary_loss_mlp": 0.0102985,
      "balance_loss_clip": 1.04319239,
      "balance_loss_mlp": 1.02151728,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.870111865849258,
      "language_loss": 0.65720624,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.67896843,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.7187747955322266
    },
    {
      "auxiliary_loss_clip": 0.01150647,
      "auxiliary_loss_mlp": 0.01028708,
      "balance_loss_clip": 1.04602873,
      "balance_loss_mlp": 1.02106667,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.075384189459096,
      "language_loss": 0.56876063,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.59055412,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.6465675830841064
    },
    {
      "auxiliary_loss_clip": 0.01157799,
      "auxiliary_loss_mlp": 0.01025911,
      "balance_loss_clip": 1.04689646,
      "balance_loss_mlp": 1.01843023,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 2.110695335484393,
      "language_loss": 0.79393083,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81576788,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.7223432064056396
    },
    {
      "auxiliary_loss_clip": 0.01150412,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 1.04772568,
      "balance_loss_mlp": 1.01633787,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.8215528063636577,
      "language_loss": 0.75693691,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.77867901,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 3.6308605670928955
    },
    {
      "auxiliary_loss_clip": 0.01164157,
      "auxiliary_loss_mlp": 0.01023783,
      "balance_loss_clip": 1.04786479,
      "balance_loss_mlp": 1.01603162,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.8531726074455286,
      "language_loss": 0.77396685,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79584622,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.583829402923584
    },
    {
      "auxiliary_loss_clip": 0.01050178,
      "auxiliary_loss_mlp": 0.0100149,
      "balance_loss_clip": 1.01225281,
      "balance_loss_mlp": 1.00057828,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6576570546157158,
      "language_loss": 0.58007598,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60059267,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.3154447078704834
    },
    {
      "auxiliary_loss_clip": 0.01159259,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04804015,
      "balance_loss_mlp": 1.0173738,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 2.3704605249024104,
      "language_loss": 0.63352549,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65537381,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.7750189304351807
    },
    {
      "auxiliary_loss_clip": 0.01173981,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.04779971,
      "balance_loss_mlp": 1.02168202,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.3210147836849946,
      "language_loss": 0.77122313,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79325885,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.6094563007354736
    },
    {
      "auxiliary_loss_clip": 0.01174187,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.05043149,
      "balance_loss_mlp": 1.01545191,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.8520101753131768,
      "language_loss": 0.81898516,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.84095919,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.6137752532958984
    },
    {
      "auxiliary_loss_clip": 0.01177322,
      "auxiliary_loss_mlp": 0.01032382,
      "balance_loss_clip": 1.052423,
      "balance_loss_mlp": 1.02443111,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.115716233237435,
      "language_loss": 0.74447381,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76657087,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 3.5380992889404297
    },
    {
      "auxiliary_loss_clip": 0.0115718,
      "auxiliary_loss_mlp": 0.01025111,
      "balance_loss_clip": 1.04669762,
      "balance_loss_mlp": 1.01802087,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 2.0071475691879894,
      "language_loss": 0.75100923,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.77283216,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.7699570655822754
    },
    {
      "auxiliary_loss_clip": 0.01117495,
      "auxiliary_loss_mlp": 0.0102811,
      "balance_loss_clip": 1.04082334,
      "balance_loss_mlp": 1.02067995,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.254314257996788,
      "language_loss": 0.64788187,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.66933793,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.7638635635375977
    },
    {
      "auxiliary_loss_clip": 0.01167996,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 1.04959118,
      "balance_loss_mlp": 1.01882887,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.3304075131015125,
      "language_loss": 0.78992617,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.81187731,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.6041295528411865
    },
    {
      "auxiliary_loss_clip": 0.01174774,
      "auxiliary_loss_mlp": 0.01028853,
      "balance_loss_clip": 1.04971814,
      "balance_loss_mlp": 1.02095532,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 2.1170339364880824,
      "language_loss": 0.73725152,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.75928777,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.5877833366394043
    },
    {
      "auxiliary_loss_clip": 0.01171851,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.05050182,
      "balance_loss_mlp": 1.01834488,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 2.1798276990786336,
      "language_loss": 0.75682479,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77880132,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.535379409790039
    },
    {
      "auxiliary_loss_clip": 0.0114771,
      "auxiliary_loss_mlp": 0.00899903,
      "balance_loss_clip": 1.0447973,
      "balance_loss_mlp": 1.00054193,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.7907876730806511,
      "language_loss": 0.74285972,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76333582,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.7601280212402344
    },
    {
      "auxiliary_loss_clip": 0.01152139,
      "auxiliary_loss_mlp": 0.01022612,
      "balance_loss_clip": 1.04600728,
      "balance_loss_mlp": 1.01498282,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 2.3913004355720053,
      "language_loss": 0.81011724,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83186471,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.6788580417633057
    },
    {
      "auxiliary_loss_clip": 0.01165391,
      "auxiliary_loss_mlp": 0.0102295,
      "balance_loss_clip": 1.04778814,
      "balance_loss_mlp": 1.0153265,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.2813893320522514,
      "language_loss": 0.72372591,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74560934,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.6425793170928955
    },
    {
      "auxiliary_loss_clip": 0.011576,
      "auxiliary_loss_mlp": 0.01031236,
      "balance_loss_clip": 1.05480742,
      "balance_loss_mlp": 1.02350271,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.8573360734124988,
      "language_loss": 0.8343209,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85620928,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 3.5444986820220947
    },
    {
      "auxiliary_loss_clip": 0.01149774,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.04503512,
      "balance_loss_mlp": 1.01768303,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.9364971190233446,
      "language_loss": 0.67610472,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69785839,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.8057572841644287
    },
    {
      "auxiliary_loss_clip": 0.01158355,
      "auxiliary_loss_mlp": 0.00901058,
      "balance_loss_clip": 1.04854524,
      "balance_loss_mlp": 1.00062394,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 3.4896145080911096,
      "language_loss": 0.8350957,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85568988,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.7167043685913086
    },
    {
      "auxiliary_loss_clip": 0.01151491,
      "auxiliary_loss_mlp": 0.01027176,
      "balance_loss_clip": 1.04746902,
      "balance_loss_mlp": 1.01957059,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.6268023748171236,
      "language_loss": 0.62759089,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.64937758,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.67484188079834
    },
    {
      "auxiliary_loss_clip": 0.01138244,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.04989767,
      "balance_loss_mlp": 1.02089703,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 2.049251313690269,
      "language_loss": 0.7527408,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77440882,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.754854202270508
    },
    {
      "auxiliary_loss_clip": 0.01154939,
      "auxiliary_loss_mlp": 0.01020915,
      "balance_loss_clip": 1.04923427,
      "balance_loss_mlp": 1.01347625,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.266436155706442,
      "language_loss": 0.859065,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.88082355,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.6290290355682373
    },
    {
      "auxiliary_loss_clip": 0.01148193,
      "auxiliary_loss_mlp": 0.01025897,
      "balance_loss_clip": 1.04399025,
      "balance_loss_mlp": 1.01821947,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.281883147298726,
      "language_loss": 0.7387718,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.76051271,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.7501821517944336
    },
    {
      "auxiliary_loss_clip": 0.0115106,
      "auxiliary_loss_mlp": 0.01025807,
      "balance_loss_clip": 1.04692984,
      "balance_loss_mlp": 1.0188477,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.9856057889427077,
      "language_loss": 0.81054497,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83231366,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.6396803855895996
    },
    {
      "auxiliary_loss_clip": 0.01129812,
      "auxiliary_loss_mlp": 0.01025134,
      "balance_loss_clip": 1.04215741,
      "balance_loss_mlp": 1.01799679,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.996178699725076,
      "language_loss": 0.75371659,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77526605,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 3.6582911014556885
    },
    {
      "auxiliary_loss_clip": 0.01167339,
      "auxiliary_loss_mlp": 0.00900343,
      "balance_loss_clip": 1.04951954,
      "balance_loss_mlp": 1.00056314,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.8858226865595757,
      "language_loss": 0.83629763,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85697448,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.723599433898926
    },
    {
      "auxiliary_loss_clip": 0.01160764,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.04830444,
      "balance_loss_mlp": 1.01747274,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.396948647038465,
      "language_loss": 0.76315016,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78500962,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.6175553798675537
    },
    {
      "auxiliary_loss_clip": 0.01167832,
      "auxiliary_loss_mlp": 0.010252,
      "balance_loss_clip": 1.04867315,
      "balance_loss_mlp": 1.01751149,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 3.160872367240886,
      "language_loss": 0.7156136,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.73754394,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.6356685161590576
    },
    {
      "auxiliary_loss_clip": 0.01146251,
      "auxiliary_loss_mlp": 0.01025165,
      "balance_loss_clip": 1.04693007,
      "balance_loss_mlp": 1.01753545,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 2.0323833638816224,
      "language_loss": 0.79129887,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81301296,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 3.5514488220214844
    },
    {
      "auxiliary_loss_clip": 0.01101984,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.03726149,
      "balance_loss_mlp": 1.01704931,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.7108510312933543,
      "language_loss": 0.76304913,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78431094,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.7524843215942383
    },
    {
      "auxiliary_loss_clip": 0.01148372,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.04684126,
      "balance_loss_mlp": 1.02053618,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.8429849618010952,
      "language_loss": 0.81962591,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.84138918,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.732187271118164
    },
    {
      "auxiliary_loss_clip": 0.01151357,
      "auxiliary_loss_mlp": 0.01024366,
      "balance_loss_clip": 1.04459023,
      "balance_loss_mlp": 1.01642084,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 3.7719069319959737,
      "language_loss": 0.86042213,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88217938,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.6775078773498535
    },
    {
      "auxiliary_loss_clip": 0.01164816,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.04926205,
      "balance_loss_mlp": 1.02260077,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.7704913634455253,
      "language_loss": 0.79410434,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81606108,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.5607781410217285
    },
    {
      "auxiliary_loss_clip": 0.01133536,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 1.04406154,
      "balance_loss_mlp": 1.01574802,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 2.10750309235116,
      "language_loss": 0.78135061,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80292118,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 3.679342031478882
    },
    {
      "auxiliary_loss_clip": 0.01158982,
      "auxiliary_loss_mlp": 0.01036945,
      "balance_loss_clip": 1.0479579,
      "balance_loss_mlp": 1.02931297,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 2.120472388795958,
      "language_loss": 0.80525148,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82721066,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.622887134552002
    },
    {
      "auxiliary_loss_clip": 0.01170287,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.05256402,
      "balance_loss_mlp": 1.01761174,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.6352893668338162,
      "language_loss": 0.57103527,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59299105,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.776841878890991
    },
    {
      "auxiliary_loss_clip": 0.01155995,
      "auxiliary_loss_mlp": 0.01028993,
      "balance_loss_clip": 1.04610562,
      "balance_loss_mlp": 1.02065444,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.9456867299713756,
      "language_loss": 0.66818702,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69003689,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.6060404777526855
    },
    {
      "auxiliary_loss_clip": 0.0117501,
      "auxiliary_loss_mlp": 0.01031913,
      "balance_loss_clip": 1.05178165,
      "balance_loss_mlp": 1.02434289,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.7518561130492176,
      "language_loss": 0.69508719,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71715641,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.5829646587371826
    },
    {
      "auxiliary_loss_clip": 0.01164955,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.04816985,
      "balance_loss_mlp": 1.02062309,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.6643406568605665,
      "language_loss": 0.72177958,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74371207,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.648149013519287
    },
    {
      "auxiliary_loss_clip": 0.01162661,
      "auxiliary_loss_mlp": 0.01025678,
      "balance_loss_clip": 1.04694486,
      "balance_loss_mlp": 1.01796484,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.6292952915051897,
      "language_loss": 0.82367224,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84555554,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.59916090965271
    },
    {
      "auxiliary_loss_clip": 0.01167795,
      "auxiliary_loss_mlp": 0.01026555,
      "balance_loss_clip": 1.04836297,
      "balance_loss_mlp": 1.01916969,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.5948074355701782,
      "language_loss": 0.70473588,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72667933,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.6493232250213623
    },
    {
      "auxiliary_loss_clip": 0.01148003,
      "auxiliary_loss_mlp": 0.01027363,
      "balance_loss_clip": 1.04670751,
      "balance_loss_mlp": 1.01995993,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.8540592605686343,
      "language_loss": 0.81918395,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.84093761,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 2.675874948501587
    },
    {
      "auxiliary_loss_clip": 0.01170143,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.05212307,
      "balance_loss_mlp": 1.01918602,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 2.3324720965981234,
      "language_loss": 0.68912685,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.7110939,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.6202447414398193
    },
    {
      "auxiliary_loss_clip": 0.01176885,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.0504266,
      "balance_loss_mlp": 1.02273703,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.0566209572678065,
      "language_loss": 0.62267166,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64474487,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 3.476749897003174
    },
    {
      "auxiliary_loss_clip": 0.01147403,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.04667509,
      "balance_loss_mlp": 1.01846302,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 2.0434992977730726,
      "language_loss": 0.75054145,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.772277,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.6380410194396973
    },
    {
      "auxiliary_loss_clip": 0.01146612,
      "auxiliary_loss_mlp": 0.01023215,
      "balance_loss_clip": 1.04598117,
      "balance_loss_mlp": 1.01650953,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.4659832534961503,
      "language_loss": 0.72672379,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.74842203,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.786116123199463
    },
    {
      "auxiliary_loss_clip": 0.01168134,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 1.04715419,
      "balance_loss_mlp": 1.01508307,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 1.7637166089333887,
      "language_loss": 0.79825115,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.82015944,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.609133005142212
    },
    {
      "auxiliary_loss_clip": 0.0107824,
      "auxiliary_loss_mlp": 0.01001795,
      "balance_loss_clip": 1.01553845,
      "balance_loss_mlp": 1.00084162,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8832076154268673,
      "language_loss": 0.60083938,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62163973,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.1524248123168945
    },
    {
      "auxiliary_loss_clip": 0.01155506,
      "auxiliary_loss_mlp": 0.01025686,
      "balance_loss_clip": 1.04560006,
      "balance_loss_mlp": 1.01769292,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6597982312425428,
      "language_loss": 0.75456607,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77637798,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.714465856552124
    },
    {
      "auxiliary_loss_clip": 0.01154016,
      "auxiliary_loss_mlp": 0.01023251,
      "balance_loss_clip": 1.04378247,
      "balance_loss_mlp": 1.0150969,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 1.9767190830600487,
      "language_loss": 0.68888533,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.71065795,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.669375419616699
    },
    {
      "auxiliary_loss_clip": 0.01174986,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.04835057,
      "balance_loss_mlp": 1.01840305,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 1.8058540886151246,
      "language_loss": 0.79677916,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.81879079,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 3.481870174407959
    },
    {
      "auxiliary_loss_clip": 0.01171413,
      "auxiliary_loss_mlp": 0.01026534,
      "balance_loss_clip": 1.04906964,
      "balance_loss_mlp": 1.01868963,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 1.8858873915671754,
      "language_loss": 0.88331401,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90529346,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.5704920291900635
    },
    {
      "auxiliary_loss_clip": 0.01158511,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.04660368,
      "balance_loss_mlp": 1.02167225,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.8367877608540024,
      "language_loss": 0.7044149,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72629023,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.6878340244293213
    },
    {
      "auxiliary_loss_clip": 0.01140855,
      "auxiliary_loss_mlp": 0.01024036,
      "balance_loss_clip": 1.04161096,
      "balance_loss_mlp": 1.01650846,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.0344256844539026,
      "language_loss": 0.60100484,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.62265378,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.698469400405884
    },
    {
      "auxiliary_loss_clip": 0.0115142,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.04683661,
      "balance_loss_mlp": 1.02088594,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.4179298193390286,
      "language_loss": 0.72403479,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74583614,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.711412191390991
    },
    {
      "auxiliary_loss_clip": 0.01118315,
      "auxiliary_loss_mlp": 0.01029386,
      "balance_loss_clip": 1.03698874,
      "balance_loss_mlp": 1.02204275,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 2.2492428067364556,
      "language_loss": 0.8587935,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.88027048,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 3.7536449432373047
    },
    {
      "auxiliary_loss_clip": 0.01159175,
      "auxiliary_loss_mlp": 0.01024518,
      "balance_loss_clip": 1.045331,
      "balance_loss_mlp": 1.01677513,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 2.8558859847446514,
      "language_loss": 0.71239686,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73423386,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.6849560737609863
    },
    {
      "auxiliary_loss_clip": 0.01159807,
      "auxiliary_loss_mlp": 0.01025393,
      "balance_loss_clip": 1.04539371,
      "balance_loss_mlp": 1.01796079,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 2.001401090073646,
      "language_loss": 0.82067561,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84252763,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.701932907104492
    },
    {
      "auxiliary_loss_clip": 0.01150904,
      "auxiliary_loss_mlp": 0.01023859,
      "balance_loss_clip": 1.04526687,
      "balance_loss_mlp": 1.01642585,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.837161993802361,
      "language_loss": 0.78227478,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80402243,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.669884443283081
    },
    {
      "auxiliary_loss_clip": 0.01154256,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.04834497,
      "balance_loss_mlp": 1.02593935,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.7623454214782541,
      "language_loss": 0.65222347,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67409879,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 3.710705518722534
    },
    {
      "auxiliary_loss_clip": 0.01171411,
      "auxiliary_loss_mlp": 0.01024521,
      "balance_loss_clip": 1.05002022,
      "balance_loss_mlp": 1.01710892,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 2.1320366137832965,
      "language_loss": 0.70022237,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.72218168,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.580170154571533
    },
    {
      "auxiliary_loss_clip": 0.01158929,
      "auxiliary_loss_mlp": 0.01029367,
      "balance_loss_clip": 1.05144823,
      "balance_loss_mlp": 1.02138567,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 4.031694079298595,
      "language_loss": 0.76247579,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.78435874,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.68837308883667
    },
    {
      "auxiliary_loss_clip": 0.01153601,
      "auxiliary_loss_mlp": 0.01024211,
      "balance_loss_clip": 1.04452157,
      "balance_loss_mlp": 1.01618814,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 4.255004943258617,
      "language_loss": 0.61166131,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63343942,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.623481273651123
    },
    {
      "auxiliary_loss_clip": 0.01158117,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04802728,
      "balance_loss_mlp": 1.01742148,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.9000336552509947,
      "language_loss": 0.90290952,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.92473626,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.594672679901123
    },
    {
      "auxiliary_loss_clip": 0.01165841,
      "auxiliary_loss_mlp": 0.0102317,
      "balance_loss_clip": 1.04751229,
      "balance_loss_mlp": 1.01580286,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 1.8153901567264696,
      "language_loss": 0.72381771,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74570787,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.703571319580078
    },
    {
      "auxiliary_loss_clip": 0.01173074,
      "auxiliary_loss_mlp": 0.01021687,
      "balance_loss_clip": 1.04962897,
      "balance_loss_mlp": 1.01412928,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 4.022735509346391,
      "language_loss": 0.86453092,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88647854,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.5335419178009033
    },
    {
      "auxiliary_loss_clip": 0.01149842,
      "auxiliary_loss_mlp": 0.01024123,
      "balance_loss_clip": 1.04217362,
      "balance_loss_mlp": 1.01602221,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 2.2067201440222064,
      "language_loss": 0.77914429,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.80088401,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.7308712005615234
    },
    {
      "auxiliary_loss_clip": 0.01137953,
      "auxiliary_loss_mlp": 0.01024363,
      "balance_loss_clip": 1.04123008,
      "balance_loss_mlp": 1.01670408,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 1.7099239094639176,
      "language_loss": 0.74796522,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.76958829,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.9183366298675537
    },
    {
      "auxiliary_loss_clip": 0.01147776,
      "auxiliary_loss_mlp": 0.01023887,
      "balance_loss_clip": 1.04611039,
      "balance_loss_mlp": 1.01597738,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.754474659921925,
      "language_loss": 0.82299626,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84471285,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.625736951828003
    },
    {
      "auxiliary_loss_clip": 0.01161801,
      "auxiliary_loss_mlp": 0.01026473,
      "balance_loss_clip": 1.04704189,
      "balance_loss_mlp": 1.01870346,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 1.8256716127560408,
      "language_loss": 0.80824924,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83013201,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 3.599574089050293
    },
    {
      "auxiliary_loss_clip": 0.01154001,
      "auxiliary_loss_mlp": 0.01020072,
      "balance_loss_clip": 1.04560387,
      "balance_loss_mlp": 1.01287723,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.7346425663182083,
      "language_loss": 0.74569881,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.7674396,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.700957775115967
    },
    {
      "auxiliary_loss_clip": 0.01163227,
      "auxiliary_loss_mlp": 0.00901034,
      "balance_loss_clip": 1.05059242,
      "balance_loss_mlp": 1.00064373,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 2.6715217204497255,
      "language_loss": 0.76878858,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.78943121,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.698483943939209
    },
    {
      "auxiliary_loss_clip": 0.01164041,
      "auxiliary_loss_mlp": 0.01023106,
      "balance_loss_clip": 1.04906738,
      "balance_loss_mlp": 1.01580489,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 2.06161255191684,
      "language_loss": 0.6956349,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71750635,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.639993190765381
    },
    {
      "auxiliary_loss_clip": 0.01160523,
      "auxiliary_loss_mlp": 0.01024424,
      "balance_loss_clip": 1.04830396,
      "balance_loss_mlp": 1.01689291,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.5072585492677106,
      "language_loss": 0.78089726,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.80274671,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.699094533920288
    },
    {
      "auxiliary_loss_clip": 0.01164361,
      "auxiliary_loss_mlp": 0.01023148,
      "balance_loss_clip": 1.04920697,
      "balance_loss_mlp": 1.01598382,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.296887446624043,
      "language_loss": 0.7678653,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78974044,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.611056089401245
    },
    {
      "auxiliary_loss_clip": 0.01172043,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.04920316,
      "balance_loss_mlp": 1.01803207,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.9001403069446223,
      "language_loss": 0.85745597,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87943113,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.675734281539917
    },
    {
      "auxiliary_loss_clip": 0.01147504,
      "auxiliary_loss_mlp": 0.01025405,
      "balance_loss_clip": 1.0419234,
      "balance_loss_mlp": 1.01818371,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.3961078911013836,
      "language_loss": 0.81940496,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.84113407,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.620720863342285
    },
    {
      "auxiliary_loss_clip": 0.01155415,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.05085945,
      "balance_loss_mlp": 1.01689756,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 1.6458528968050503,
      "language_loss": 0.67148304,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69328481,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 3.706198215484619
    },
    {
      "auxiliary_loss_clip": 0.01130356,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04537868,
      "balance_loss_mlp": 1.01908875,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.7048089295153803,
      "language_loss": 0.78197706,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80355483,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.771296739578247
    },
    {
      "auxiliary_loss_clip": 0.01141413,
      "auxiliary_loss_mlp": 0.01022913,
      "balance_loss_clip": 1.04017687,
      "balance_loss_mlp": 1.01542091,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 2.1407538225489198,
      "language_loss": 0.78148502,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.8031283,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.714494466781616
    },
    {
      "auxiliary_loss_clip": 0.01147585,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 1.04458833,
      "balance_loss_mlp": 1.01973701,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.27094964757797,
      "language_loss": 0.71226656,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73401415,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.684481143951416
    },
    {
      "auxiliary_loss_clip": 0.0115733,
      "auxiliary_loss_mlp": 0.01022876,
      "balance_loss_clip": 1.045259,
      "balance_loss_mlp": 1.01580119,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 2.0428509066833014,
      "language_loss": 0.85636127,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87816334,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.5873308181762695
    },
    {
      "auxiliary_loss_clip": 0.01177898,
      "auxiliary_loss_mlp": 0.01023032,
      "balance_loss_clip": 1.05188394,
      "balance_loss_mlp": 1.01574278,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 2.3349383878409666,
      "language_loss": 0.72187853,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.7438879,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 3.686814785003662
    },
    {
      "auxiliary_loss_clip": 0.0113447,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 1.04132879,
      "balance_loss_mlp": 1.01507998,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 2.3814434240050812,
      "language_loss": 0.76268089,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78425419,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.7198801040649414
    },
    {
      "auxiliary_loss_clip": 0.01174918,
      "auxiliary_loss_mlp": 0.00901101,
      "balance_loss_clip": 1.05071449,
      "balance_loss_mlp": 1.00068665,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 1.9566631916716817,
      "language_loss": 0.82542646,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.84618664,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.5812878608703613
    },
    {
      "auxiliary_loss_clip": 0.01162941,
      "auxiliary_loss_mlp": 0.01023673,
      "balance_loss_clip": 1.04849589,
      "balance_loss_mlp": 1.01591206,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.351760073280585,
      "language_loss": 0.79830199,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82016814,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.616874933242798
    },
    {
      "auxiliary_loss_clip": 0.01166781,
      "auxiliary_loss_mlp": 0.01023385,
      "balance_loss_clip": 1.05139422,
      "balance_loss_mlp": 1.01634836,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 7.084900631559246,
      "language_loss": 0.83590269,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85780442,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 3.631312608718872
    },
    {
      "auxiliary_loss_clip": 0.01165604,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.04737389,
      "balance_loss_mlp": 1.02329421,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.8759340598501444,
      "language_loss": 0.84347934,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.86543608,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.6393818855285645
    },
    {
      "auxiliary_loss_clip": 0.01162524,
      "auxiliary_loss_mlp": 0.01026067,
      "balance_loss_clip": 1.04813755,
      "balance_loss_mlp": 1.01859879,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 2.418950092089234,
      "language_loss": 0.78505927,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.8069452,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.6175107955932617
    },
    {
      "auxiliary_loss_clip": 0.01140868,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.04626906,
      "balance_loss_mlp": 1.01996875,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.5567321427257257,
      "language_loss": 0.74318194,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76486856,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.709381580352783
    },
    {
      "auxiliary_loss_clip": 0.01152602,
      "auxiliary_loss_mlp": 0.0102199,
      "balance_loss_clip": 1.04524863,
      "balance_loss_mlp": 1.01430964,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 2.0554767501930815,
      "language_loss": 0.65903091,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68077683,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.706970453262329
    },
    {
      "auxiliary_loss_clip": 0.01145022,
      "auxiliary_loss_mlp": 0.01024388,
      "balance_loss_clip": 1.04937136,
      "balance_loss_mlp": 1.01670814,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.7075826128252123,
      "language_loss": 0.79898548,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82067961,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.719305992126465
    },
    {
      "auxiliary_loss_clip": 0.01073975,
      "auxiliary_loss_mlp": 0.01000993,
      "balance_loss_clip": 1.01644754,
      "balance_loss_mlp": 1.00005138,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.835386261126386,
      "language_loss": 0.55651444,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57726413,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.1930699348449707
    },
    {
      "auxiliary_loss_clip": 0.0115732,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04484892,
      "balance_loss_mlp": 1.01740003,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.8430699918546842,
      "language_loss": 0.75150877,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77333486,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.6330490112304688
    },
    {
      "auxiliary_loss_clip": 0.0117212,
      "auxiliary_loss_mlp": 0.01021875,
      "balance_loss_clip": 1.04960716,
      "balance_loss_mlp": 1.0143882,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 1.8701217755402288,
      "language_loss": 0.712129,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73406893,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 2.6282198429107666
    },
    {
      "auxiliary_loss_clip": 0.01151866,
      "auxiliary_loss_mlp": 0.01023871,
      "balance_loss_clip": 1.04467666,
      "balance_loss_mlp": 1.01620317,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 3.19041963951933,
      "language_loss": 0.6923365,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71409386,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 3.7513370513916016
    },
    {
      "auxiliary_loss_clip": 0.01160631,
      "auxiliary_loss_mlp": 0.0103031,
      "balance_loss_clip": 1.04758048,
      "balance_loss_mlp": 1.02241874,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.313039770072592,
      "language_loss": 0.82155645,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.84346581,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.6459388732910156
    },
    {
      "auxiliary_loss_clip": 0.01153404,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04722512,
      "balance_loss_mlp": 1.0184288,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.8572521267179405,
      "language_loss": 0.81362379,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83541358,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.605177640914917
    },
    {
      "auxiliary_loss_clip": 0.01163287,
      "auxiliary_loss_mlp": 0.01024235,
      "balance_loss_clip": 1.04963255,
      "balance_loss_mlp": 1.01733923,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.5233719602513887,
      "language_loss": 0.83822131,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.86009657,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.6365091800689697
    },
    {
      "auxiliary_loss_clip": 0.01138177,
      "auxiliary_loss_mlp": 0.01022596,
      "balance_loss_clip": 1.0480057,
      "balance_loss_mlp": 1.01472199,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 1.879983476730091,
      "language_loss": 0.73665035,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.7582581,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.682742118835449
    },
    {
      "auxiliary_loss_clip": 0.01141117,
      "auxiliary_loss_mlp": 0.01030401,
      "balance_loss_clip": 1.04474497,
      "balance_loss_mlp": 1.02228236,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.1813688310026293,
      "language_loss": 0.64925492,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67097008,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.726069688796997
    },
    {
      "auxiliary_loss_clip": 0.01159472,
      "auxiliary_loss_mlp": 0.01021932,
      "balance_loss_clip": 1.04873109,
      "balance_loss_mlp": 1.0143857,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 1.6970021166828697,
      "language_loss": 0.80415571,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82596976,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.6995019912719727
    },
    {
      "auxiliary_loss_clip": 0.01163851,
      "auxiliary_loss_mlp": 0.01029718,
      "balance_loss_clip": 1.04683411,
      "balance_loss_mlp": 1.02195764,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.096734635474265,
      "language_loss": 0.8337869,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85572261,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.6576292514801025
    },
    {
      "auxiliary_loss_clip": 0.01069168,
      "auxiliary_loss_mlp": 0.01007593,
      "balance_loss_clip": 1.03054833,
      "balance_loss_mlp": 1.0065974,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6932781761748608,
      "language_loss": 0.49596763,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51673526,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.3101935386657715
    },
    {
      "auxiliary_loss_clip": 0.01127791,
      "auxiliary_loss_mlp": 0.01030217,
      "balance_loss_clip": 1.0429002,
      "balance_loss_mlp": 1.02270103,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 13.776289308942571,
      "language_loss": 0.71875906,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74033916,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 3.6626672744750977
    },
    {
      "auxiliary_loss_clip": 0.01172477,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.04905212,
      "balance_loss_mlp": 1.01885724,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.86596542950367,
      "language_loss": 0.8814401,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90342796,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.5399909019470215
    },
    {
      "auxiliary_loss_clip": 0.01145715,
      "auxiliary_loss_mlp": 0.01024188,
      "balance_loss_clip": 1.04408169,
      "balance_loss_mlp": 1.01734257,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.759116510216817,
      "language_loss": 0.74264789,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76434696,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.650294065475464
    },
    {
      "auxiliary_loss_clip": 0.01148293,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.04725575,
      "balance_loss_mlp": 1.01839471,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.9175927617260402,
      "language_loss": 0.71723497,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.738976,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.6963398456573486
    },
    {
      "auxiliary_loss_clip": 0.01144889,
      "auxiliary_loss_mlp": 0.01027781,
      "balance_loss_clip": 1.04634225,
      "balance_loss_mlp": 1.02056849,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 1.811146615314377,
      "language_loss": 0.68555546,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70728219,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 3.667318105697632
    },
    {
      "auxiliary_loss_clip": 0.01061232,
      "auxiliary_loss_mlp": 0.01000608,
      "balance_loss_clip": 1.01531112,
      "balance_loss_mlp": 0.99968451,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.8431812640444296,
      "language_loss": 0.5670256,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58764398,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.2687952518463135
    },
    {
      "auxiliary_loss_clip": 0.01159492,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.05360484,
      "balance_loss_mlp": 1.01638818,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.365772354303968,
      "language_loss": 0.76957124,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79141009,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.6662042140960693
    },
    {
      "auxiliary_loss_clip": 0.01160727,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.04613698,
      "balance_loss_mlp": 1.02004802,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.7515805464303085,
      "language_loss": 0.72954202,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75142515,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.688192129135132
    },
    {
      "auxiliary_loss_clip": 0.01146934,
      "auxiliary_loss_mlp": 0.01021645,
      "balance_loss_clip": 1.0432117,
      "balance_loss_mlp": 1.0144093,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.356693997557804,
      "language_loss": 0.78791481,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80960059,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 3.7244713306427
    },
    {
      "auxiliary_loss_clip": 0.01164815,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04823136,
      "balance_loss_mlp": 1.01852453,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 1.848895699127594,
      "language_loss": 0.7221812,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74409354,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.659067392349243
    },
    {
      "auxiliary_loss_clip": 0.01117135,
      "auxiliary_loss_mlp": 0.00901176,
      "balance_loss_clip": 1.03977203,
      "balance_loss_mlp": 1.00074291,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.7113333981790808,
      "language_loss": 0.75805533,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77823842,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.7332308292388916
    },
    {
      "auxiliary_loss_clip": 0.01166659,
      "auxiliary_loss_mlp": 0.01019441,
      "balance_loss_clip": 1.04767156,
      "balance_loss_mlp": 1.01215696,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.9412478680584366,
      "language_loss": 0.652964,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67482495,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.6276636123657227
    },
    {
      "auxiliary_loss_clip": 0.01174205,
      "auxiliary_loss_mlp": 0.01026853,
      "balance_loss_clip": 1.05047524,
      "balance_loss_mlp": 1.01871705,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.8812417828441381,
      "language_loss": 0.85633832,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87834889,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.5802719593048096
    },
    {
      "auxiliary_loss_clip": 0.01150378,
      "auxiliary_loss_mlp": 0.01021893,
      "balance_loss_clip": 1.04717016,
      "balance_loss_mlp": 1.01487422,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.946176696195625,
      "language_loss": 0.83333707,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85505974,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.7364132404327393
    },
    {
      "auxiliary_loss_clip": 0.01160035,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.0475682,
      "balance_loss_mlp": 1.01650143,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 2.287501908845457,
      "language_loss": 0.66103458,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68287134,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.6107099056243896
    },
    {
      "auxiliary_loss_clip": 0.01076337,
      "auxiliary_loss_mlp": 0.01000761,
      "balance_loss_clip": 1.01402116,
      "balance_loss_mlp": 0.99984342,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0102019045223694,
      "language_loss": 0.67731476,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69808578,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 2.9453625679016113
    },
    {
      "auxiliary_loss_clip": 0.0114495,
      "auxiliary_loss_mlp": 0.01022962,
      "balance_loss_clip": 1.04577363,
      "balance_loss_mlp": 1.01552391,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.8149987003419121,
      "language_loss": 0.76135635,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78303546,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.752856492996216
    },
    {
      "auxiliary_loss_clip": 0.01072659,
      "auxiliary_loss_mlp": 0.01003783,
      "balance_loss_clip": 1.01543546,
      "balance_loss_mlp": 1.002841,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7403135406314488,
      "language_loss": 0.56321043,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58397484,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.0895209312438965
    },
    {
      "auxiliary_loss_clip": 0.01161319,
      "auxiliary_loss_mlp": 0.01021209,
      "balance_loss_clip": 1.04412043,
      "balance_loss_mlp": 1.01480484,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 1.8134388833065824,
      "language_loss": 0.85704005,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87886536,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 3.5708205699920654
    },
    {
      "auxiliary_loss_clip": 0.01148678,
      "auxiliary_loss_mlp": 0.01021211,
      "balance_loss_clip": 1.04559767,
      "balance_loss_mlp": 1.01361132,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 7.257070995729064,
      "language_loss": 0.70066983,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72236878,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.6488192081451416
    },
    {
      "auxiliary_loss_clip": 0.01170306,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04795349,
      "balance_loss_mlp": 1.01825404,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 2.0739948505104686,
      "language_loss": 0.77385247,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79580986,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.5774006843566895
    },
    {
      "auxiliary_loss_clip": 0.01152106,
      "auxiliary_loss_mlp": 0.01026064,
      "balance_loss_clip": 1.04425812,
      "balance_loss_mlp": 1.01796353,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.8611498809545415,
      "language_loss": 0.8126111,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83439279,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.614189624786377
    },
    {
      "auxiliary_loss_clip": 0.0115853,
      "auxiliary_loss_mlp": 0.01030652,
      "balance_loss_clip": 1.04846144,
      "balance_loss_mlp": 1.02307343,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.7110091533436136,
      "language_loss": 0.71611249,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73800439,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.647735357284546
    },
    {
      "auxiliary_loss_clip": 0.01159331,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.04541564,
      "balance_loss_mlp": 1.01896882,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 1.9438844719113437,
      "language_loss": 0.73982966,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.76168215,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.648930311203003
    },
    {
      "auxiliary_loss_clip": 0.01158219,
      "auxiliary_loss_mlp": 0.01023363,
      "balance_loss_clip": 1.04661095,
      "balance_loss_mlp": 1.0162611,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 2.054050639323031,
      "language_loss": 0.73290521,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75472105,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.6537926197052
    },
    {
      "auxiliary_loss_clip": 0.0115248,
      "auxiliary_loss_mlp": 0.01026437,
      "balance_loss_clip": 1.04550648,
      "balance_loss_mlp": 1.01929927,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.2615184090715705,
      "language_loss": 0.80155516,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82334429,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.641604423522949
    },
    {
      "auxiliary_loss_clip": 0.01156468,
      "auxiliary_loss_mlp": 0.01031979,
      "balance_loss_clip": 1.0465765,
      "balance_loss_mlp": 1.02418244,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 1.7137851638704935,
      "language_loss": 0.85128081,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87316531,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.622309684753418
    },
    {
      "auxiliary_loss_clip": 0.01160067,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 1.04539549,
      "balance_loss_mlp": 1.02254951,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.6781942645985217,
      "language_loss": 0.73103946,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.75294662,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 3.906154155731201
    },
    {
      "auxiliary_loss_clip": 0.01172269,
      "auxiliary_loss_mlp": 0.01028699,
      "balance_loss_clip": 1.05217898,
      "balance_loss_mlp": 1.02097464,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 2.747026136857506,
      "language_loss": 0.75239795,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77440763,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.601045608520508
    },
    {
      "auxiliary_loss_clip": 0.01152378,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 1.05098951,
      "balance_loss_mlp": 1.01905382,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.960167094792195,
      "language_loss": 0.79493791,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81672561,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.7093091011047363
    },
    {
      "auxiliary_loss_clip": 0.01162937,
      "auxiliary_loss_mlp": 0.01024017,
      "balance_loss_clip": 1.04796648,
      "balance_loss_mlp": 1.01663494,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 1.9905861993925178,
      "language_loss": 0.70786822,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.72973776,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.657426595687866
    },
    {
      "auxiliary_loss_clip": 0.01149817,
      "auxiliary_loss_mlp": 0.01022265,
      "balance_loss_clip": 1.04475999,
      "balance_loss_mlp": 1.01535964,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.5441121185594266,
      "language_loss": 0.82393324,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84565401,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 3.6731534004211426
    },
    {
      "auxiliary_loss_clip": 0.01160549,
      "auxiliary_loss_mlp": 0.01025756,
      "balance_loss_clip": 1.04664862,
      "balance_loss_mlp": 1.01859128,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 2.6064890408859163,
      "language_loss": 0.71954989,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.741413,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.6315884590148926
    },
    {
      "auxiliary_loss_clip": 0.01155313,
      "auxiliary_loss_mlp": 0.01019864,
      "balance_loss_clip": 1.04536796,
      "balance_loss_mlp": 1.01269364,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.2765379665343484,
      "language_loss": 0.72728413,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.74903584,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.60866641998291
    },
    {
      "auxiliary_loss_clip": 0.01145379,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.04427266,
      "balance_loss_mlp": 1.01661849,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.6888040630037882,
      "language_loss": 0.76060545,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78229564,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.6788227558135986
    },
    {
      "auxiliary_loss_clip": 0.01161697,
      "auxiliary_loss_mlp": 0.01027859,
      "balance_loss_clip": 1.04503191,
      "balance_loss_mlp": 1.02031338,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.719129923622191,
      "language_loss": 0.76504493,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78694052,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 3.6392924785614014
    },
    {
      "auxiliary_loss_clip": 0.01154898,
      "auxiliary_loss_mlp": 0.01021579,
      "balance_loss_clip": 1.04430771,
      "balance_loss_mlp": 1.01414585,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 3.0103805609402436,
      "language_loss": 0.7542696,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.77603441,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.649631977081299
    },
    {
      "auxiliary_loss_clip": 0.01163065,
      "auxiliary_loss_mlp": 0.01021824,
      "balance_loss_clip": 1.04607999,
      "balance_loss_mlp": 1.0141474,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 1.95104128267889,
      "language_loss": 0.75776374,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.77961266,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.694553852081299
    },
    {
      "auxiliary_loss_clip": 0.01135235,
      "auxiliary_loss_mlp": 0.01023378,
      "balance_loss_clip": 1.03984988,
      "balance_loss_mlp": 1.01617169,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.3072214587717914,
      "language_loss": 0.86121297,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88279903,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.752431869506836
    },
    {
      "auxiliary_loss_clip": 0.01162262,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 1.0451045,
      "balance_loss_mlp": 1.01880217,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.8761243993688854,
      "language_loss": 0.6902228,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71210963,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.6854381561279297
    },
    {
      "auxiliary_loss_clip": 0.01158286,
      "auxiliary_loss_mlp": 0.01023249,
      "balance_loss_clip": 1.0451088,
      "balance_loss_mlp": 1.0158397,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.680788383453657,
      "language_loss": 0.69556433,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71737969,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.6424317359924316
    },
    {
      "auxiliary_loss_clip": 0.01158863,
      "auxiliary_loss_mlp": 0.00901313,
      "balance_loss_clip": 1.04681003,
      "balance_loss_mlp": 1.00080156,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 5.372390532765539,
      "language_loss": 0.81854832,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.83915013,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.7165064811706543
    },
    {
      "auxiliary_loss_clip": 0.01154098,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.04688334,
      "balance_loss_mlp": 1.02072883,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 1.9196991389679758,
      "language_loss": 0.77449423,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79631996,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.7259881496429443
    },
    {
      "auxiliary_loss_clip": 0.01158378,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.04356968,
      "balance_loss_mlp": 1.01972246,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 2.0630425826214807,
      "language_loss": 0.81953681,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.84139204,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.633409023284912
    },
    {
      "auxiliary_loss_clip": 0.01125861,
      "auxiliary_loss_mlp": 0.0090045,
      "balance_loss_clip": 1.04124486,
      "balance_loss_mlp": 1.00048447,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 3.3121567502360065,
      "language_loss": 0.64789706,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.6681602,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.8011679649353027
    },
    {
      "auxiliary_loss_clip": 0.01153237,
      "auxiliary_loss_mlp": 0.01023049,
      "balance_loss_clip": 1.04578471,
      "balance_loss_mlp": 1.01609051,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 1.795097404554418,
      "language_loss": 0.81807339,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83983624,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 3.6584019660949707
    },
    {
      "auxiliary_loss_clip": 0.01159267,
      "auxiliary_loss_mlp": 0.01029925,
      "balance_loss_clip": 1.04854107,
      "balance_loss_mlp": 1.02152705,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 1.9505287808728735,
      "language_loss": 0.80989367,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.83178556,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.6026246547698975
    },
    {
      "auxiliary_loss_clip": 0.01158242,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04654539,
      "balance_loss_mlp": 1.01524556,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 3.573946590047905,
      "language_loss": 0.75591826,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77772969,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.6179842948913574
    },
    {
      "auxiliary_loss_clip": 0.01142629,
      "auxiliary_loss_mlp": 0.01024106,
      "balance_loss_clip": 1.04590392,
      "balance_loss_mlp": 1.01687264,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.9707115532418107,
      "language_loss": 0.72140491,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74307227,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.646604061126709
    },
    {
      "auxiliary_loss_clip": 0.01156373,
      "auxiliary_loss_mlp": 0.01019556,
      "balance_loss_clip": 1.04407382,
      "balance_loss_mlp": 1.01228476,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.8493409022456557,
      "language_loss": 0.70759755,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72935688,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.6379268169403076
    },
    {
      "auxiliary_loss_clip": 0.01154066,
      "auxiliary_loss_mlp": 0.01030173,
      "balance_loss_clip": 1.0457108,
      "balance_loss_mlp": 1.02280855,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 3.1883595968220555,
      "language_loss": 0.7867583,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80860078,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.6392626762390137
    },
    {
      "auxiliary_loss_clip": 0.01149316,
      "auxiliary_loss_mlp": 0.0102898,
      "balance_loss_clip": 1.0442636,
      "balance_loss_mlp": 1.02146351,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.7386013741679824,
      "language_loss": 0.75916111,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.78094411,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.648164987564087
    },
    {
      "auxiliary_loss_clip": 0.01162908,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.05194497,
      "balance_loss_mlp": 1.02127051,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 2.063966797651247,
      "language_loss": 0.79092354,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.81284511,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.726646661758423
    },
    {
      "auxiliary_loss_clip": 0.01158345,
      "auxiliary_loss_mlp": 0.01029419,
      "balance_loss_clip": 1.04684162,
      "balance_loss_mlp": 1.02244806,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 2.3014930030553544,
      "language_loss": 0.7977047,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.81958234,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 3.548170328140259
    },
    {
      "auxiliary_loss_clip": 0.01163197,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.04990697,
      "balance_loss_mlp": 1.01711774,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.5772587155412126,
      "language_loss": 0.76971424,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79158854,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.6497702598571777
    },
    {
      "auxiliary_loss_clip": 0.01151681,
      "auxiliary_loss_mlp": 0.01023806,
      "balance_loss_clip": 1.0464747,
      "balance_loss_mlp": 1.01643264,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 3.6438435536999183,
      "language_loss": 0.73057967,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75233454,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.6900622844696045
    },
    {
      "auxiliary_loss_clip": 0.01152092,
      "auxiliary_loss_mlp": 0.0102254,
      "balance_loss_clip": 1.04727376,
      "balance_loss_mlp": 1.01535773,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.4512439763399034,
      "language_loss": 0.79812264,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.81986892,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.676149845123291
    },
    {
      "auxiliary_loss_clip": 0.01154189,
      "auxiliary_loss_mlp": 0.01022209,
      "balance_loss_clip": 1.0450418,
      "balance_loss_mlp": 1.01499116,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 2.6158174020342604,
      "language_loss": 0.83291906,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85468304,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.6172850131988525
    },
    {
      "auxiliary_loss_clip": 0.01065725,
      "auxiliary_loss_mlp": 0.01001734,
      "balance_loss_clip": 1.01179886,
      "balance_loss_mlp": 1.00085759,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7495926063316334,
      "language_loss": 0.56091833,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58159292,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 4.084993839263916
    },
    {
      "auxiliary_loss_clip": 0.01126002,
      "auxiliary_loss_mlp": 0.0102163,
      "balance_loss_clip": 1.03940034,
      "balance_loss_mlp": 1.01438236,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.7193796849962286,
      "language_loss": 0.76496816,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78644454,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.778383731842041
    },
    {
      "auxiliary_loss_clip": 0.0114893,
      "auxiliary_loss_mlp": 0.01022052,
      "balance_loss_clip": 1.04502296,
      "balance_loss_mlp": 1.01488161,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 2.187332982562146,
      "language_loss": 0.61242688,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63413668,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.6659321784973145
    },
    {
      "auxiliary_loss_clip": 0.0114522,
      "auxiliary_loss_mlp": 0.01030589,
      "balance_loss_clip": 1.04536176,
      "balance_loss_mlp": 1.02271819,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.9019827893783785,
      "language_loss": 0.68657631,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70833439,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.701087474822998
    },
    {
      "auxiliary_loss_clip": 0.01168331,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.04863834,
      "balance_loss_mlp": 1.01888967,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 1.9619292808644342,
      "language_loss": 0.6588341,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.68078172,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 3.6160545349121094
    },
    {
      "auxiliary_loss_clip": 0.01168861,
      "auxiliary_loss_mlp": 0.01023678,
      "balance_loss_clip": 1.04886127,
      "balance_loss_mlp": 1.01582181,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.0334521217460875,
      "language_loss": 0.69623435,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71815968,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.655426502227783
    },
    {
      "auxiliary_loss_clip": 0.01165144,
      "auxiliary_loss_mlp": 0.01025331,
      "balance_loss_clip": 1.04491591,
      "balance_loss_mlp": 1.01789224,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 3.1502228078541785,
      "language_loss": 0.81567448,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83757925,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.705599784851074
    },
    {
      "auxiliary_loss_clip": 0.01060376,
      "auxiliary_loss_mlp": 0.01000432,
      "balance_loss_clip": 1.01617455,
      "balance_loss_mlp": 0.99941885,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8176199380804606,
      "language_loss": 0.61960828,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64021635,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.3465330600738525
    },
    {
      "auxiliary_loss_clip": 0.01163237,
      "auxiliary_loss_mlp": 0.01021985,
      "balance_loss_clip": 1.04931974,
      "balance_loss_mlp": 1.01433444,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.2896401269157614,
      "language_loss": 0.70242202,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.72427422,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.6142594814300537
    },
    {
      "auxiliary_loss_clip": 0.01155076,
      "auxiliary_loss_mlp": 0.01028192,
      "balance_loss_clip": 1.04636562,
      "balance_loss_mlp": 1.02109337,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.9644475695283152,
      "language_loss": 0.72718948,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74902213,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.6445300579071045
    },
    {
      "auxiliary_loss_clip": 0.01069813,
      "auxiliary_loss_mlp": 0.01000196,
      "balance_loss_clip": 1.01234651,
      "balance_loss_mlp": 0.99934405,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7712732335144299,
      "language_loss": 0.61016095,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63086104,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.1395742893218994
    },
    {
      "auxiliary_loss_clip": 0.01138991,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.04409909,
      "balance_loss_mlp": 1.020298,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.2131323646881245,
      "language_loss": 0.84986347,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87153703,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.71362042427063
    },
    {
      "auxiliary_loss_clip": 0.0117294,
      "auxiliary_loss_mlp": 0.01031879,
      "balance_loss_clip": 1.04795992,
      "balance_loss_mlp": 1.02445793,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 2.8814787652909613,
      "language_loss": 0.789204,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81125224,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.560687780380249
    },
    {
      "auxiliary_loss_clip": 0.01170735,
      "auxiliary_loss_mlp": 0.01026656,
      "balance_loss_clip": 1.04693651,
      "balance_loss_mlp": 1.01894295,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.295175803872647,
      "language_loss": 0.80002314,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82199705,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.586656093597412
    },
    {
      "auxiliary_loss_clip": 0.01134697,
      "auxiliary_loss_mlp": 0.00900943,
      "balance_loss_clip": 1.0446918,
      "balance_loss_mlp": 1.00074005,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.3588662481330287,
      "language_loss": 0.71557784,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73593426,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 3.664440393447876
    },
    {
      "auxiliary_loss_clip": 0.01139101,
      "auxiliary_loss_mlp": 0.01023169,
      "balance_loss_clip": 1.04403925,
      "balance_loss_mlp": 1.01619232,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 2.0458016691598178,
      "language_loss": 0.71425807,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73588073,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.6685004234313965
    },
    {
      "auxiliary_loss_clip": 0.0117559,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.05396712,
      "balance_loss_mlp": 1.02368891,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.0649509288761743,
      "language_loss": 0.82433748,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84639943,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.598001480102539
    },
    {
      "auxiliary_loss_clip": 0.01143636,
      "auxiliary_loss_mlp": 0.00901331,
      "balance_loss_clip": 1.04141128,
      "balance_loss_mlp": 1.00065088,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.3319632398482057,
      "language_loss": 0.77014363,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.79059339,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.633207082748413
    },
    {
      "auxiliary_loss_clip": 0.01151609,
      "auxiliary_loss_mlp": 0.01031622,
      "balance_loss_clip": 1.04823875,
      "balance_loss_mlp": 1.02370667,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.288823489035554,
      "language_loss": 0.81371421,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83554649,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.740922451019287
    },
    {
      "auxiliary_loss_clip": 0.01162503,
      "auxiliary_loss_mlp": 0.01025875,
      "balance_loss_clip": 1.04828238,
      "balance_loss_mlp": 1.01837683,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.2501271791204234,
      "language_loss": 0.76912844,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.79101217,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.5911006927490234
    },
    {
      "auxiliary_loss_clip": 0.01159791,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.04700327,
      "balance_loss_mlp": 1.01836729,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 1.80367769891032,
      "language_loss": 0.72776663,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74962634,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.709582567214966
    },
    {
      "auxiliary_loss_clip": 0.01153124,
      "auxiliary_loss_mlp": 0.010229,
      "balance_loss_clip": 1.04574895,
      "balance_loss_mlp": 1.01556897,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 2.0440983870886624,
      "language_loss": 0.74456728,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76632756,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 3.5667459964752197
    },
    {
      "auxiliary_loss_clip": 0.01141797,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.04360068,
      "balance_loss_mlp": 1.02027357,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 2.050076432244283,
      "language_loss": 0.83488798,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.85658753,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.6595983505249023
    },
    {
      "auxiliary_loss_clip": 0.01169919,
      "auxiliary_loss_mlp": 0.01024028,
      "balance_loss_clip": 1.04795361,
      "balance_loss_mlp": 1.01675272,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 2.7004944859662636,
      "language_loss": 0.76804423,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.78998363,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.5996434688568115
    },
    {
      "auxiliary_loss_clip": 0.01122201,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.03937244,
      "balance_loss_mlp": 1.0179956,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.7775670970803175,
      "language_loss": 0.71720105,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73867816,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.7226402759552
    },
    {
      "auxiliary_loss_clip": 0.01123213,
      "auxiliary_loss_mlp": 0.01027731,
      "balance_loss_clip": 1.04316819,
      "balance_loss_mlp": 1.01935673,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 3.5771240146683345,
      "language_loss": 0.64635158,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.66786104,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.8956165313720703
    },
    {
      "auxiliary_loss_clip": 0.01158688,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.05000639,
      "balance_loss_mlp": 1.02252519,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.4856165181336751,
      "language_loss": 0.64037621,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66226375,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 4.015719175338745
    },
    {
      "auxiliary_loss_clip": 0.0116212,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04859424,
      "balance_loss_mlp": 1.0184021,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 1.9413156555714501,
      "language_loss": 0.69089746,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71278107,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.6152169704437256
    },
    {
      "auxiliary_loss_clip": 0.01133452,
      "auxiliary_loss_mlp": 0.01029512,
      "balance_loss_clip": 1.04419935,
      "balance_loss_mlp": 1.021662,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.7215754751958288,
      "language_loss": 0.80654252,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82817209,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.662130355834961
    },
    {
      "auxiliary_loss_clip": 0.01159412,
      "auxiliary_loss_mlp": 0.01023719,
      "balance_loss_clip": 1.04455447,
      "balance_loss_mlp": 1.01591074,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 2.5286903175879987,
      "language_loss": 0.75396764,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.77579892,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.6595187187194824
    },
    {
      "auxiliary_loss_clip": 0.01149692,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.04513669,
      "balance_loss_mlp": 1.01659966,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 1.9772115396171552,
      "language_loss": 0.77491224,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.79665345,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 3.7252037525177
    },
    {
      "auxiliary_loss_clip": 0.0116976,
      "auxiliary_loss_mlp": 0.01023396,
      "balance_loss_clip": 1.0480988,
      "balance_loss_mlp": 1.01583207,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.7882840777107274,
      "language_loss": 0.76584047,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78777206,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.6266212463378906
    },
    {
      "auxiliary_loss_clip": 0.01154855,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.05120146,
      "balance_loss_mlp": 1.0203011,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 2.0132837639540466,
      "language_loss": 0.77636617,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.79819536,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.666311740875244
    },
    {
      "auxiliary_loss_clip": 0.01155286,
      "auxiliary_loss_mlp": 0.0102322,
      "balance_loss_clip": 1.04565752,
      "balance_loss_mlp": 1.01575136,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.9057041881287013,
      "language_loss": 0.72784936,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.74963444,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.5653131008148193
    },
    {
      "auxiliary_loss_clip": 0.01068999,
      "auxiliary_loss_mlp": 0.01010829,
      "balance_loss_clip": 1.03012705,
      "balance_loss_mlp": 1.00992286,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8301175830226395,
      "language_loss": 0.54995453,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.57075274,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.2196779251098633
    },
    {
      "auxiliary_loss_clip": 0.01132198,
      "auxiliary_loss_mlp": 0.01027889,
      "balance_loss_clip": 1.04481101,
      "balance_loss_mlp": 1.02058184,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.2068077447967793,
      "language_loss": 0.81143838,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83303928,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.718534469604492
    },
    {
      "auxiliary_loss_clip": 0.01140613,
      "auxiliary_loss_mlp": 0.00900952,
      "balance_loss_clip": 1.0447855,
      "balance_loss_mlp": 1.00067306,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 2.151001800289187,
      "language_loss": 0.71888983,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.7393055,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.7198028564453125
    },
    {
      "auxiliary_loss_clip": 0.01167312,
      "auxiliary_loss_mlp": 0.01025828,
      "balance_loss_clip": 1.04920685,
      "balance_loss_mlp": 1.01770353,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 2.0522406988622643,
      "language_loss": 0.8172158,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.83914727,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.701561212539673
    },
    {
      "auxiliary_loss_clip": 0.0117886,
      "auxiliary_loss_mlp": 0.01029147,
      "balance_loss_clip": 1.05082273,
      "balance_loss_mlp": 1.02124357,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.142814416555996,
      "language_loss": 0.70032185,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72240198,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.679229259490967
    },
    {
      "auxiliary_loss_clip": 0.01135154,
      "auxiliary_loss_mlp": 0.01022433,
      "balance_loss_clip": 1.04362583,
      "balance_loss_mlp": 1.0154531,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.7873930588525824,
      "language_loss": 0.81588101,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83745688,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.7748141288757324
    },
    {
      "auxiliary_loss_clip": 0.01158541,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.04614115,
      "balance_loss_mlp": 1.01479149,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.7349406775569134,
      "language_loss": 0.62511325,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64692259,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 3.8823916912078857
    },
    {
      "auxiliary_loss_clip": 0.01168655,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.04675281,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 2.1592503269189245,
      "language_loss": 0.75378513,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77571875,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.5816330909729004
    },
    {
      "auxiliary_loss_clip": 0.0106157,
      "auxiliary_loss_mlp": 0.01001429,
      "balance_loss_clip": 1.01225364,
      "balance_loss_mlp": 1.00062394,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7768292226993019,
      "language_loss": 0.58577543,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60640544,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.314034938812256
    },
    {
      "auxiliary_loss_clip": 0.01147087,
      "auxiliary_loss_mlp": 0.0102407,
      "balance_loss_clip": 1.0449729,
      "balance_loss_mlp": 1.01661658,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.654931978907627,
      "language_loss": 0.68149745,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70320904,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.725994348526001
    },
    {
      "auxiliary_loss_clip": 0.01153016,
      "auxiliary_loss_mlp": 0.01022005,
      "balance_loss_clip": 1.04507208,
      "balance_loss_mlp": 1.01533222,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 3.3720018682817083,
      "language_loss": 0.82107323,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84282345,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.6567776203155518
    },
    {
      "auxiliary_loss_clip": 0.0115613,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.04352272,
      "balance_loss_mlp": 1.02100694,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.318234282783793,
      "language_loss": 0.79581773,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81766677,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.6250767707824707
    },
    {
      "auxiliary_loss_clip": 0.01173902,
      "auxiliary_loss_mlp": 0.01024753,
      "balance_loss_clip": 1.05092978,
      "balance_loss_mlp": 1.01721287,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.7788295391229088,
      "language_loss": 0.81325889,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83524537,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.591074228286743
    },
    {
      "auxiliary_loss_clip": 0.01163397,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.04966307,
      "balance_loss_mlp": 1.01871443,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 2.044869931230376,
      "language_loss": 0.89150298,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91339511,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.558814525604248
    },
    {
      "auxiliary_loss_clip": 0.01174122,
      "auxiliary_loss_mlp": 0.01023119,
      "balance_loss_clip": 1.04810619,
      "balance_loss_mlp": 1.01551056,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 1.9827347000811029,
      "language_loss": 0.70550549,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72747791,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 3.5342283248901367
    },
    {
      "auxiliary_loss_clip": 0.01175262,
      "auxiliary_loss_mlp": 0.01021412,
      "balance_loss_clip": 1.0495882,
      "balance_loss_mlp": 1.01406264,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.9865212687061935,
      "language_loss": 0.70086873,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72283548,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.5474002361297607
    },
    {
      "auxiliary_loss_clip": 0.01146347,
      "auxiliary_loss_mlp": 0.01025929,
      "balance_loss_clip": 1.04472947,
      "balance_loss_mlp": 1.01870775,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 2.080514055387078,
      "language_loss": 0.6794461,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70116884,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.6315135955810547
    },
    {
      "auxiliary_loss_clip": 0.01161776,
      "auxiliary_loss_mlp": 0.01023814,
      "balance_loss_clip": 1.04688454,
      "balance_loss_mlp": 1.01625562,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.8064321562768866,
      "language_loss": 0.80236852,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82422441,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.5924508571624756
    },
    {
      "auxiliary_loss_clip": 0.01161028,
      "auxiliary_loss_mlp": 0.01025192,
      "balance_loss_clip": 1.05094242,
      "balance_loss_mlp": 1.01887655,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.8815297513673312,
      "language_loss": 0.84017974,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86204195,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 3.7661216259002686
    },
    {
      "auxiliary_loss_clip": 0.01148523,
      "auxiliary_loss_mlp": 0.00900979,
      "balance_loss_clip": 1.04473948,
      "balance_loss_mlp": 1.00066113,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.7783843583739156,
      "language_loss": 0.80389994,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82439494,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.686375617980957
    },
    {
      "auxiliary_loss_clip": 0.01172479,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.04953206,
      "balance_loss_mlp": 1.01919937,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 1.8129931794611807,
      "language_loss": 0.80997777,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83196861,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.5539727210998535
    },
    {
      "auxiliary_loss_clip": 0.01156927,
      "auxiliary_loss_mlp": 0.01022171,
      "balance_loss_clip": 1.04531157,
      "balance_loss_mlp": 1.01492262,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.5228500179218807,
      "language_loss": 0.73749614,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.75928712,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.690819263458252
    },
    {
      "auxiliary_loss_clip": 0.01170016,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04744446,
      "balance_loss_mlp": 1.02002943,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 1.9565798962379854,
      "language_loss": 0.6691972,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.69117403,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.5316007137298584
    },
    {
      "auxiliary_loss_clip": 0.01160757,
      "auxiliary_loss_mlp": 0.01028888,
      "balance_loss_clip": 1.04587483,
      "balance_loss_mlp": 1.020841,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.6010614025790075,
      "language_loss": 0.74117303,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.76306951,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 3.5225980281829834
    },
    {
      "auxiliary_loss_clip": 0.01167812,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04654622,
      "balance_loss_mlp": 1.02000213,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.9262462114354157,
      "language_loss": 0.76324987,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78520173,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.584402084350586
    },
    {
      "auxiliary_loss_clip": 0.01171741,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.0498718,
      "balance_loss_mlp": 1.01817441,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 3.3388686212589738,
      "language_loss": 0.6888538,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.71083105,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.638298749923706
    },
    {
      "auxiliary_loss_clip": 0.01171074,
      "auxiliary_loss_mlp": 0.01028268,
      "balance_loss_clip": 1.04805434,
      "balance_loss_mlp": 1.02091265,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.6541713719709474,
      "language_loss": 0.83887875,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.86087215,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.5639286041259766
    },
    {
      "auxiliary_loss_clip": 0.01143804,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.04128373,
      "balance_loss_mlp": 1.01734185,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.8235067564456398,
      "language_loss": 0.76019728,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78188634,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.7467241287231445
    },
    {
      "auxiliary_loss_clip": 0.01156576,
      "auxiliary_loss_mlp": 0.01026549,
      "balance_loss_clip": 1.04685497,
      "balance_loss_mlp": 1.01926541,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 2.7193655775701617,
      "language_loss": 0.79199082,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81382209,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.640366792678833
    },
    {
      "auxiliary_loss_clip": 0.01147931,
      "auxiliary_loss_mlp": 0.01028726,
      "balance_loss_clip": 1.04544425,
      "balance_loss_mlp": 1.02128744,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 4.05843033037349,
      "language_loss": 0.80285597,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82462257,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.7029309272766113
    },
    {
      "auxiliary_loss_clip": 0.01134371,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04095685,
      "balance_loss_mlp": 1.0206337,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 2.166902166222226,
      "language_loss": 0.81747401,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83910358,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.7167928218841553
    },
    {
      "auxiliary_loss_clip": 0.01160801,
      "auxiliary_loss_mlp": 0.01023263,
      "balance_loss_clip": 1.04647207,
      "balance_loss_mlp": 1.01577985,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 2.1521901457157813,
      "language_loss": 0.77117264,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.79301327,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.551278829574585
    },
    {
      "auxiliary_loss_clip": 0.0116116,
      "auxiliary_loss_mlp": 0.01024387,
      "balance_loss_clip": 1.04924238,
      "balance_loss_mlp": 1.01667416,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.8002155605372439,
      "language_loss": 0.80054134,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82239676,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 3.6695661544799805
    },
    {
      "auxiliary_loss_clip": 0.01150587,
      "auxiliary_loss_mlp": 0.01025913,
      "balance_loss_clip": 1.04775345,
      "balance_loss_mlp": 1.01845598,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.836975205893935,
      "language_loss": 0.72619134,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.74795634,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.707235097885132
    },
    {
      "auxiliary_loss_clip": 0.01153683,
      "auxiliary_loss_mlp": 0.01025394,
      "balance_loss_clip": 1.04324293,
      "balance_loss_mlp": 1.01753807,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.650914035484139,
      "language_loss": 0.64072007,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66251081,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.7554891109466553
    },
    {
      "auxiliary_loss_clip": 0.01128666,
      "auxiliary_loss_mlp": 0.01023913,
      "balance_loss_clip": 1.04353523,
      "balance_loss_mlp": 1.01681077,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 2.1878765584681177,
      "language_loss": 0.82172513,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.84325093,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.838028907775879
    },
    {
      "auxiliary_loss_clip": 0.01055218,
      "auxiliary_loss_mlp": 0.0100325,
      "balance_loss_clip": 1.01297843,
      "balance_loss_mlp": 1.00235045,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9250534752891015,
      "language_loss": 0.67239976,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69298446,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.1535937786102295
    },
    {
      "auxiliary_loss_clip": 0.0117732,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.05132782,
      "balance_loss_mlp": 1.02020061,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.899572308163501,
      "language_loss": 0.68259102,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70464659,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.650665283203125
    },
    {
      "auxiliary_loss_clip": 0.0113197,
      "auxiliary_loss_mlp": 0.01022671,
      "balance_loss_clip": 1.04275167,
      "balance_loss_mlp": 1.01581347,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.9450982094017728,
      "language_loss": 0.82607955,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84762597,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.6813199520111084
    },
    {
      "auxiliary_loss_clip": 0.01154801,
      "auxiliary_loss_mlp": 0.01031431,
      "balance_loss_clip": 1.04570353,
      "balance_loss_mlp": 1.02340198,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 2.094279462112918,
      "language_loss": 0.75596279,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.77782512,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.748239278793335
    },
    {
      "auxiliary_loss_clip": 0.01161882,
      "auxiliary_loss_mlp": 0.01020824,
      "balance_loss_clip": 1.04629731,
      "balance_loss_mlp": 1.01431513,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 2.1314909172082284,
      "language_loss": 0.83284879,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85467583,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 2.5955138206481934
    },
    {
      "auxiliary_loss_clip": 0.01070445,
      "auxiliary_loss_mlp": 0.01000633,
      "balance_loss_clip": 1.01256561,
      "balance_loss_mlp": 0.99969703,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7126642849838872,
      "language_loss": 0.54679775,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56750852,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 4.185089826583862
    },
    {
      "auxiliary_loss_clip": 0.01129054,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 1.0441494,
      "balance_loss_mlp": 1.01926351,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.013049528607054,
      "language_loss": 0.71471554,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73627257,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.686952590942383
    },
    {
      "auxiliary_loss_clip": 0.01155798,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.04690206,
      "balance_loss_mlp": 1.01954937,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.3424989724907066,
      "language_loss": 0.68653172,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.708363,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.5943210124969482
    },
    {
      "auxiliary_loss_clip": 0.0116471,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.04730082,
      "balance_loss_mlp": 1.0194484,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.5338920485846472,
      "language_loss": 0.82590389,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.8478297,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.600925922393799
    },
    {
      "auxiliary_loss_clip": 0.01173355,
      "auxiliary_loss_mlp": 0.01032438,
      "balance_loss_clip": 1.04877949,
      "balance_loss_mlp": 1.02491021,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 2.4089519264492507,
      "language_loss": 0.72153378,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74359179,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 3.5114035606384277
    },
    {
      "auxiliary_loss_clip": 0.01145125,
      "auxiliary_loss_mlp": 0.01024204,
      "balance_loss_clip": 1.04586124,
      "balance_loss_mlp": 1.01673508,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.6722883589574018,
      "language_loss": 0.76885867,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.7905519,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.6558682918548584
    },
    {
      "auxiliary_loss_clip": 0.01147506,
      "auxiliary_loss_mlp": 0.01028924,
      "balance_loss_clip": 1.04281759,
      "balance_loss_mlp": 1.02130675,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 2.121841793648093,
      "language_loss": 0.68976188,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71152622,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.749021053314209
    },
    {
      "auxiliary_loss_clip": 0.01165045,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.0490458,
      "balance_loss_mlp": 1.01676583,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 7.631165016117777,
      "language_loss": 0.73523772,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75713307,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.5778603553771973
    },
    {
      "auxiliary_loss_clip": 0.0115745,
      "auxiliary_loss_mlp": 0.01021896,
      "balance_loss_clip": 1.04490495,
      "balance_loss_mlp": 1.01475191,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 1.9422486400114711,
      "language_loss": 0.87886119,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.90065467,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.6888468265533447
    },
    {
      "auxiliary_loss_clip": 0.01168951,
      "auxiliary_loss_mlp": 0.01021535,
      "balance_loss_clip": 1.04897475,
      "balance_loss_mlp": 1.01512456,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.9870683708716896,
      "language_loss": 0.73190069,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75380558,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 3.523625135421753
    },
    {
      "auxiliary_loss_clip": 0.01158971,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04643309,
      "balance_loss_mlp": 1.01756144,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 2.0337806194877066,
      "language_loss": 0.64277166,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66461468,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.6610794067382812
    },
    {
      "auxiliary_loss_clip": 0.01147591,
      "auxiliary_loss_mlp": 0.0102791,
      "balance_loss_clip": 1.04569459,
      "balance_loss_mlp": 1.02087653,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.8446529090321921,
      "language_loss": 0.81616026,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.8379153,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.7560956478118896
    },
    {
      "auxiliary_loss_clip": 0.01155004,
      "auxiliary_loss_mlp": 0.01020757,
      "balance_loss_clip": 1.0454874,
      "balance_loss_mlp": 1.01327991,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 2.219073192139467,
      "language_loss": 0.7441107,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76586831,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.624377727508545
    },
    {
      "auxiliary_loss_clip": 0.01157986,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.0463407,
      "balance_loss_mlp": 1.01624322,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 3.173786803572936,
      "language_loss": 0.7488243,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.77064538,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.6556851863861084
    },
    {
      "auxiliary_loss_clip": 0.01137085,
      "auxiliary_loss_mlp": 0.01020207,
      "balance_loss_clip": 1.04133368,
      "balance_loss_mlp": 1.01213598,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 6.743515669833773,
      "language_loss": 0.77739817,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.79897106,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.673358917236328
    },
    {
      "auxiliary_loss_clip": 0.01136335,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.0438118,
      "balance_loss_mlp": 1.01939738,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 2.208016998609167,
      "language_loss": 0.66614038,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68777329,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.6781113147735596
    },
    {
      "auxiliary_loss_clip": 0.01167951,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.04882097,
      "balance_loss_mlp": 1.02574539,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 2.8367116677911404,
      "language_loss": 0.68590236,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.7079221,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.581468343734741
    },
    {
      "auxiliary_loss_clip": 0.01133162,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.04198146,
      "balance_loss_mlp": 1.02106094,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 1.9523049533937018,
      "language_loss": 0.66671622,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.6883359,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.850274085998535
    },
    {
      "auxiliary_loss_clip": 0.01136266,
      "auxiliary_loss_mlp": 0.01022661,
      "balance_loss_clip": 1.04411077,
      "balance_loss_mlp": 1.01597023,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.9580682454922576,
      "language_loss": 0.8205772,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84216654,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 3.5835959911346436
    },
    {
      "auxiliary_loss_clip": 0.01154292,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.0480988,
      "balance_loss_mlp": 1.0197531,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.018599394716289,
      "language_loss": 0.78199518,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80381036,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.724130868911743
    },
    {
      "auxiliary_loss_clip": 0.01159406,
      "auxiliary_loss_mlp": 0.00900883,
      "balance_loss_clip": 1.04775727,
      "balance_loss_mlp": 1.00064087,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.4981969828990505,
      "language_loss": 0.84075159,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.86135447,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.712144613265991
    },
    {
      "auxiliary_loss_clip": 0.01165595,
      "auxiliary_loss_mlp": 0.01027626,
      "balance_loss_clip": 1.0484035,
      "balance_loss_mlp": 1.02033949,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 2.1353044400258514,
      "language_loss": 0.73222417,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75415635,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.608098268508911
    },
    {
      "auxiliary_loss_clip": 0.01069884,
      "auxiliary_loss_mlp": 0.01001106,
      "balance_loss_clip": 1.0130012,
      "balance_loss_mlp": 1.00027168,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8131714890457007,
      "language_loss": 0.54579473,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56650466,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.0726189613342285
    },
    {
      "auxiliary_loss_clip": 0.01155456,
      "auxiliary_loss_mlp": 0.01022774,
      "balance_loss_clip": 1.04831398,
      "balance_loss_mlp": 1.01614308,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.97455372801381,
      "language_loss": 0.56383854,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58562088,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.651296854019165
    },
    {
      "auxiliary_loss_clip": 0.01149862,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.04721332,
      "balance_loss_mlp": 1.01732755,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 2.1000292119521373,
      "language_loss": 0.76148367,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78322399,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.7308125495910645
    },
    {
      "auxiliary_loss_clip": 0.01147069,
      "auxiliary_loss_mlp": 0.00901081,
      "balance_loss_clip": 1.04272342,
      "balance_loss_mlp": 1.00062776,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 2.396035906112456,
      "language_loss": 0.81542504,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.83590651,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 2.7234041690826416
    },
    {
      "auxiliary_loss_clip": 0.01160191,
      "auxiliary_loss_mlp": 0.01024734,
      "balance_loss_clip": 1.04680741,
      "balance_loss_mlp": 1.01749182,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 3.712776553714178,
      "language_loss": 0.6864332,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70828241,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.6771600246429443
    },
    {
      "auxiliary_loss_clip": 0.01151633,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.05270648,
      "balance_loss_mlp": 1.02051961,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 1.690119637054692,
      "language_loss": 0.81416082,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83595651,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 3.537578582763672
    },
    {
      "auxiliary_loss_clip": 0.01066267,
      "auxiliary_loss_mlp": 0.01000632,
      "balance_loss_clip": 1.01244366,
      "balance_loss_mlp": 0.99976158,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8468931988743279,
      "language_loss": 0.65265471,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67332369,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.1548964977264404
    },
    {
      "auxiliary_loss_clip": 0.01142166,
      "auxiliary_loss_mlp": 0.01023471,
      "balance_loss_clip": 1.04730392,
      "balance_loss_mlp": 1.01600218,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 2.054693413696822,
      "language_loss": 0.74692011,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.7685765,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 2.8015568256378174
    },
    {
      "auxiliary_loss_clip": 0.01162423,
      "auxiliary_loss_mlp": 0.01021049,
      "balance_loss_clip": 1.04535747,
      "balance_loss_mlp": 1.01341343,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.5288372994310393,
      "language_loss": 0.74067056,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76250535,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.631831169128418
    },
    {
      "auxiliary_loss_clip": 0.01155817,
      "auxiliary_loss_mlp": 0.0090148,
      "balance_loss_clip": 1.04380369,
      "balance_loss_mlp": 1.00068903,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.9839354285145092,
      "language_loss": 0.8160581,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.83663106,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 3.5402047634124756
    },
    {
      "auxiliary_loss_clip": 0.0115223,
      "auxiliary_loss_mlp": 0.01025644,
      "balance_loss_clip": 1.04228354,
      "balance_loss_mlp": 1.01845598,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.927015637986544,
      "language_loss": 0.74264002,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76441884,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.6545259952545166
    },
    {
      "auxiliary_loss_clip": 0.01169168,
      "auxiliary_loss_mlp": 0.01031616,
      "balance_loss_clip": 1.04850066,
      "balance_loss_mlp": 1.02382541,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.424897991074855,
      "language_loss": 0.74410677,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76611459,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.58134126663208
    },
    {
      "auxiliary_loss_clip": 0.01155797,
      "auxiliary_loss_mlp": 0.0102775,
      "balance_loss_clip": 1.04979026,
      "balance_loss_mlp": 1.02057946,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.8235656030069018,
      "language_loss": 0.68212116,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.7039566,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.7016232013702393
    },
    {
      "auxiliary_loss_clip": 0.01137289,
      "auxiliary_loss_mlp": 0.01026283,
      "balance_loss_clip": 1.04358804,
      "balance_loss_mlp": 1.01944959,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.7518521073841418,
      "language_loss": 0.82281733,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.8444531,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.6494059562683105
    },
    {
      "auxiliary_loss_clip": 0.01131285,
      "auxiliary_loss_mlp": 0.01020694,
      "balance_loss_clip": 1.04080367,
      "balance_loss_mlp": 1.01269555,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.9732143571129286,
      "language_loss": 0.66028923,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68180901,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 3.6494081020355225
    },
    {
      "auxiliary_loss_clip": 0.01170704,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.04566646,
      "balance_loss_mlp": 1.01977706,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.088104109403701,
      "language_loss": 0.62227201,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64425766,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.546297788619995
    },
    {
      "auxiliary_loss_clip": 0.01174325,
      "auxiliary_loss_mlp": 0.01029071,
      "balance_loss_clip": 1.04922485,
      "balance_loss_mlp": 1.0215075,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.653508544901914,
      "language_loss": 0.70086539,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.72289944,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.5643458366394043
    },
    {
      "auxiliary_loss_clip": 0.0115306,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04467475,
      "balance_loss_mlp": 1.01601326,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.7372924610132463,
      "language_loss": 0.69846189,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.72023124,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.5542986392974854
    },
    {
      "auxiliary_loss_clip": 0.01139096,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.04424286,
      "balance_loss_mlp": 1.02238142,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 5.962743695679511,
      "language_loss": 0.84319556,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.8648904,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.7957730293273926
    },
    {
      "auxiliary_loss_clip": 0.01155691,
      "auxiliary_loss_mlp": 0.01023225,
      "balance_loss_clip": 1.04544687,
      "balance_loss_mlp": 1.01578021,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.395241102053125,
      "language_loss": 0.83683866,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85862792,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.676544189453125
    },
    {
      "auxiliary_loss_clip": 0.01161082,
      "auxiliary_loss_mlp": 0.00900889,
      "balance_loss_clip": 1.04861629,
      "balance_loss_mlp": 1.00067139,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 2.0416864868834415,
      "language_loss": 0.84241283,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86303258,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.612696409225464
    },
    {
      "auxiliary_loss_clip": 0.01157055,
      "auxiliary_loss_mlp": 0.00901029,
      "balance_loss_clip": 1.04509413,
      "balance_loss_mlp": 1.00068116,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 4.299675201609965,
      "language_loss": 0.64145577,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.6620366,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.7399468421936035
    },
    {
      "auxiliary_loss_clip": 0.01161269,
      "auxiliary_loss_mlp": 0.00901202,
      "balance_loss_clip": 1.04961061,
      "balance_loss_mlp": 1.00079226,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 2.4189323144515265,
      "language_loss": 0.7994802,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82010496,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.630530595779419
    },
    {
      "auxiliary_loss_clip": 0.01157679,
      "auxiliary_loss_mlp": 0.01024038,
      "balance_loss_clip": 1.04990077,
      "balance_loss_mlp": 1.01599717,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.179339807564664,
      "language_loss": 0.76218021,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78399736,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 3.636073350906372
    },
    {
      "auxiliary_loss_clip": 0.01147382,
      "auxiliary_loss_mlp": 0.01022459,
      "balance_loss_clip": 1.04408169,
      "balance_loss_mlp": 1.01527071,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.6053668300536768,
      "language_loss": 0.85094023,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87263864,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.6719093322753906
    },
    {
      "auxiliary_loss_clip": 0.0114904,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04520202,
      "balance_loss_mlp": 1.0180825,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.0682620901610735,
      "language_loss": 0.71496922,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73671627,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.714571237564087
    },
    {
      "auxiliary_loss_clip": 0.01159713,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.04547369,
      "balance_loss_mlp": 1.02043533,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.9345747286555055,
      "language_loss": 0.70304048,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72491097,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.6187217235565186
    },
    {
      "auxiliary_loss_clip": 0.01151952,
      "auxiliary_loss_mlp": 0.01024866,
      "balance_loss_clip": 1.04474151,
      "balance_loss_mlp": 1.01767206,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.8519100193162368,
      "language_loss": 0.58033288,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60210109,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.7838351726531982
    },
    {
      "auxiliary_loss_clip": 0.01144535,
      "auxiliary_loss_mlp": 0.01029395,
      "balance_loss_clip": 1.05077195,
      "balance_loss_mlp": 1.02193832,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.7871400879649477,
      "language_loss": 0.70268166,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72442091,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.7634682655334473
    },
    {
      "auxiliary_loss_clip": 0.01147337,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.0509522,
      "balance_loss_mlp": 1.02125919,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 5.867874900216439,
      "language_loss": 0.7172128,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73897231,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.702040672302246
    },
    {
      "auxiliary_loss_clip": 0.01152811,
      "auxiliary_loss_mlp": 0.01025037,
      "balance_loss_clip": 1.04801226,
      "balance_loss_mlp": 1.01666522,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.6816895151703386,
      "language_loss": 0.77754235,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79932082,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.785011053085327
    },
    {
      "auxiliary_loss_clip": 0.01163439,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.04526663,
      "balance_loss_mlp": 1.02042115,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.7352435372453885,
      "language_loss": 0.78992647,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81183821,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 3.548708915710449
    },
    {
      "auxiliary_loss_clip": 0.01174511,
      "auxiliary_loss_mlp": 0.01031286,
      "balance_loss_clip": 1.05102396,
      "balance_loss_mlp": 1.02350736,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 2.1034581000419603,
      "language_loss": 0.75532335,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.7773813,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.559438705444336
    },
    {
      "auxiliary_loss_clip": 0.01159162,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.04835403,
      "balance_loss_mlp": 1.02003682,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 2.1130601652135317,
      "language_loss": 0.85034037,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87220025,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.5674288272857666
    },
    {
      "auxiliary_loss_clip": 0.01164621,
      "auxiliary_loss_mlp": 0.00901341,
      "balance_loss_clip": 1.04931116,
      "balance_loss_mlp": 1.00073695,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 2.0191477138386036,
      "language_loss": 0.84219813,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.8628577,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.5973405838012695
    },
    {
      "auxiliary_loss_clip": 0.0115953,
      "auxiliary_loss_mlp": 0.01029062,
      "balance_loss_clip": 1.05034208,
      "balance_loss_mlp": 1.02131331,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.8942894552552882,
      "language_loss": 0.83093774,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.85282362,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 3.6155314445495605
    },
    {
      "auxiliary_loss_clip": 0.01149647,
      "auxiliary_loss_mlp": 0.01022289,
      "balance_loss_clip": 1.04562676,
      "balance_loss_mlp": 1.01442719,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 1.8846195809857555,
      "language_loss": 0.71646798,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.73818731,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.6756458282470703
    },
    {
      "auxiliary_loss_clip": 0.0116207,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.04990387,
      "balance_loss_mlp": 1.01731443,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 2.010485573004201,
      "language_loss": 0.78174835,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80361646,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.78590726852417
    },
    {
      "auxiliary_loss_clip": 0.01177701,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.05041075,
      "balance_loss_mlp": 1.01760209,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 2.4911660298512106,
      "language_loss": 0.72273606,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74476874,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.535825252532959
    },
    {
      "auxiliary_loss_clip": 0.01127883,
      "auxiliary_loss_mlp": 0.00901732,
      "balance_loss_clip": 1.04265213,
      "balance_loss_mlp": 1.00061321,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 2.7948625909186795,
      "language_loss": 0.7619803,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78227651,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.759489059448242
    },
    {
      "auxiliary_loss_clip": 0.01173016,
      "auxiliary_loss_mlp": 0.01023344,
      "balance_loss_clip": 1.0508703,
      "balance_loss_mlp": 1.0161767,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 2.5701087846328474,
      "language_loss": 0.86637968,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88834333,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.6118431091308594
    },
    {
      "auxiliary_loss_clip": 0.01167881,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 1.04833555,
      "balance_loss_mlp": 1.02003455,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 4.647031333421381,
      "language_loss": 0.7568624,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77881366,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 3.5503172874450684
    },
    {
      "auxiliary_loss_clip": 0.01169926,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.04678416,
      "balance_loss_mlp": 1.0168885,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.4401546280624147,
      "language_loss": 0.74418032,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76612711,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.603285551071167
    },
    {
      "auxiliary_loss_clip": 0.01163461,
      "auxiliary_loss_mlp": 0.01022179,
      "balance_loss_clip": 1.0446291,
      "balance_loss_mlp": 1.01463926,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.8085606685845996,
      "language_loss": 0.78566468,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.8075211,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.628725290298462
    },
    {
      "auxiliary_loss_clip": 0.01134236,
      "auxiliary_loss_mlp": 0.01026587,
      "balance_loss_clip": 1.04683518,
      "balance_loss_mlp": 1.01839113,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.7684853752619165,
      "language_loss": 0.70689249,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72850066,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.778757095336914
    },
    {
      "auxiliary_loss_clip": 0.01160119,
      "auxiliary_loss_mlp": 0.00901049,
      "balance_loss_clip": 1.04564548,
      "balance_loss_mlp": 1.00062859,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 2.517637194452186,
      "language_loss": 0.65139353,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67200518,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.670356512069702
    },
    {
      "auxiliary_loss_clip": 0.01164383,
      "auxiliary_loss_mlp": 0.01024092,
      "balance_loss_clip": 1.04721248,
      "balance_loss_mlp": 1.01670694,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.9355072462720373,
      "language_loss": 0.60425878,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62614357,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.623972177505493
    },
    {
      "auxiliary_loss_clip": 0.01158759,
      "auxiliary_loss_mlp": 0.01022636,
      "balance_loss_clip": 1.04710937,
      "balance_loss_mlp": 1.01495314,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 3.955734020708906,
      "language_loss": 0.75016701,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.771981,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.571340322494507
    },
    {
      "auxiliary_loss_clip": 0.01165004,
      "auxiliary_loss_mlp": 0.01028762,
      "balance_loss_clip": 1.05323148,
      "balance_loss_mlp": 1.02045953,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.3855290576543053,
      "language_loss": 0.7426008,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76453853,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.6348934173583984
    },
    {
      "auxiliary_loss_clip": 0.01158713,
      "auxiliary_loss_mlp": 0.01023381,
      "balance_loss_clip": 1.04630017,
      "balance_loss_mlp": 1.01582265,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.606897554132492,
      "language_loss": 0.73885351,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76067448,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 3.672673463821411
    },
    {
      "auxiliary_loss_clip": 0.0105604,
      "auxiliary_loss_mlp": 0.01003512,
      "balance_loss_clip": 1.01498783,
      "balance_loss_mlp": 1.00259459,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8769089360973354,
      "language_loss": 0.61692679,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63752222,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.346911668777466
    },
    {
      "auxiliary_loss_clip": 0.01171303,
      "auxiliary_loss_mlp": 0.01026466,
      "balance_loss_clip": 1.0483191,
      "balance_loss_mlp": 1.0188005,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.9557193653860137,
      "language_loss": 0.91737592,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.9393537,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.6194872856140137
    },
    {
      "auxiliary_loss_clip": 0.01065933,
      "auxiliary_loss_mlp": 0.01003055,
      "balance_loss_clip": 1.01427841,
      "balance_loss_mlp": 1.00221467,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7586854154464027,
      "language_loss": 0.58963716,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.61032712,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.3326058387756348
    },
    {
      "auxiliary_loss_clip": 0.0112616,
      "auxiliary_loss_mlp": 0.01020989,
      "balance_loss_clip": 1.04516554,
      "balance_loss_mlp": 1.01335347,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.8904414924607702,
      "language_loss": 0.67069072,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69216222,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.8600478172302246
    },
    {
      "auxiliary_loss_clip": 0.01068776,
      "auxiliary_loss_mlp": 0.0100259,
      "balance_loss_clip": 1.01209795,
      "balance_loss_mlp": 1.00181508,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9184347287412764,
      "language_loss": 0.71343386,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73414755,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.1548357009887695
    },
    {
      "auxiliary_loss_clip": 0.01164085,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.04800272,
      "balance_loss_mlp": 1.02230024,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.835408693557743,
      "language_loss": 0.7289592,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.75089884,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 2.641570806503296
    },
    {
      "auxiliary_loss_clip": 0.01172252,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.04999447,
      "balance_loss_mlp": 1.01438308,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 2.0340744378489877,
      "language_loss": 0.81900179,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84094214,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.569453477859497
    },
    {
      "auxiliary_loss_clip": 0.01138034,
      "auxiliary_loss_mlp": 0.01031743,
      "balance_loss_clip": 1.04465246,
      "balance_loss_mlp": 1.02336907,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.2201045918274853,
      "language_loss": 0.7531262,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77482402,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.7257931232452393
    },
    {
      "auxiliary_loss_clip": 0.01151405,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.04536104,
      "balance_loss_mlp": 1.0186255,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 3.5471268981022748,
      "language_loss": 0.70089859,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72267389,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 3.535235643386841
    },
    {
      "auxiliary_loss_clip": 0.01063731,
      "auxiliary_loss_mlp": 0.00890658,
      "balance_loss_clip": 1.01077902,
      "balance_loss_mlp": 1.00002432,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7849025683072877,
      "language_loss": 0.55680782,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.5763517,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 3.071516990661621
    },
    {
      "auxiliary_loss_clip": 0.01075823,
      "auxiliary_loss_mlp": 0.01009263,
      "balance_loss_clip": 1.03935027,
      "balance_loss_mlp": 1.00821972,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7770241444897588,
      "language_loss": 0.63681078,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65766168,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.141770601272583
    },
    {
      "auxiliary_loss_clip": 0.01157837,
      "auxiliary_loss_mlp": 0.01024906,
      "balance_loss_clip": 1.04620349,
      "balance_loss_mlp": 1.01742005,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.8289725488144004,
      "language_loss": 0.77920711,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80103463,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 3.534714460372925
    },
    {
      "auxiliary_loss_clip": 0.0116373,
      "auxiliary_loss_mlp": 0.00900816,
      "balance_loss_clip": 1.04826093,
      "balance_loss_mlp": 1.00068748,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 1.8779882222855493,
      "language_loss": 0.68724525,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70789069,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.6074752807617188
    },
    {
      "auxiliary_loss_clip": 0.01148299,
      "auxiliary_loss_mlp": 0.01027297,
      "balance_loss_clip": 1.05331218,
      "balance_loss_mlp": 1.02034163,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 1.6788328155165733,
      "language_loss": 0.71631658,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73807251,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.8277475833892822
    },
    {
      "auxiliary_loss_clip": 0.01149446,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.04272676,
      "balance_loss_mlp": 1.01988626,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.8815171097413406,
      "language_loss": 0.85446024,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87622386,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.6457433700561523
    },
    {
      "auxiliary_loss_clip": 0.01138537,
      "auxiliary_loss_mlp": 0.01030699,
      "balance_loss_clip": 1.04711628,
      "balance_loss_mlp": 1.0223875,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.37081102915152,
      "language_loss": 0.7384634,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.76015574,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.735346555709839
    },
    {
      "auxiliary_loss_clip": 0.01146712,
      "auxiliary_loss_mlp": 0.01027266,
      "balance_loss_clip": 1.05059981,
      "balance_loss_mlp": 1.01973844,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.999453579078321,
      "language_loss": 0.64234453,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66408432,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.62919282913208
    },
    {
      "auxiliary_loss_clip": 0.01136642,
      "auxiliary_loss_mlp": 0.00901754,
      "balance_loss_clip": 1.04635322,
      "balance_loss_mlp": 1.0006206,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.3771864791736372,
      "language_loss": 0.76540422,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78578812,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 3.5889463424682617
    },
    {
      "auxiliary_loss_clip": 0.01161921,
      "auxiliary_loss_mlp": 0.01026716,
      "balance_loss_clip": 1.04672718,
      "balance_loss_mlp": 1.01948333,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.0006719567422384,
      "language_loss": 0.76063502,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78252137,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.5973939895629883
    },
    {
      "auxiliary_loss_clip": 0.01160907,
      "auxiliary_loss_mlp": 0.01022676,
      "balance_loss_clip": 1.04756439,
      "balance_loss_mlp": 1.0154103,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.5008295118569217,
      "language_loss": 0.75805116,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77988696,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.6209194660186768
    },
    {
      "auxiliary_loss_clip": 0.01173435,
      "auxiliary_loss_mlp": 0.01020049,
      "balance_loss_clip": 1.0498538,
      "balance_loss_mlp": 1.01265812,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 2.044528864874965,
      "language_loss": 0.75748563,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77942044,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.6575522422790527
    },
    {
      "auxiliary_loss_clip": 0.01163143,
      "auxiliary_loss_mlp": 0.01033558,
      "balance_loss_clip": 1.0445261,
      "balance_loss_mlp": 1.0249573,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 2.670942481734266,
      "language_loss": 0.7273398,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.7493068,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.5845890045166016
    },
    {
      "auxiliary_loss_clip": 0.01169669,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 1.04892039,
      "balance_loss_mlp": 1.01906729,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.898913755887344,
      "language_loss": 0.75945318,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78141272,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.639127016067505
    },
    {
      "auxiliary_loss_clip": 0.01151779,
      "auxiliary_loss_mlp": 0.01024998,
      "balance_loss_clip": 1.04543519,
      "balance_loss_mlp": 1.01761639,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.7458648758765591,
      "language_loss": 0.83690459,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.85867226,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.775007963180542
    },
    {
      "auxiliary_loss_clip": 0.01167306,
      "auxiliary_loss_mlp": 0.01025914,
      "balance_loss_clip": 1.04934406,
      "balance_loss_mlp": 1.01853502,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.823920149954395,
      "language_loss": 0.75356007,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77549231,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.708401679992676
    },
    {
      "auxiliary_loss_clip": 0.01069578,
      "auxiliary_loss_mlp": 0.01001529,
      "balance_loss_clip": 1.01287615,
      "balance_loss_mlp": 1.00067079,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.843637608279206,
      "language_loss": 0.58549869,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60620975,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.220916271209717
    },
    {
      "auxiliary_loss_clip": 0.01166065,
      "auxiliary_loss_mlp": 0.01021929,
      "balance_loss_clip": 1.05283141,
      "balance_loss_mlp": 1.01525354,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.6636829741041272,
      "language_loss": 0.77961385,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80149388,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 3.5705809593200684
    },
    {
      "auxiliary_loss_clip": 0.01138593,
      "auxiliary_loss_mlp": 0.00900853,
      "balance_loss_clip": 1.04170966,
      "balance_loss_mlp": 1.00074887,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.5576490550435171,
      "language_loss": 0.71998262,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.74037713,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.860426425933838
    },
    {
      "auxiliary_loss_clip": 0.0114505,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.04227996,
      "balance_loss_mlp": 1.01956117,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.586129993510874,
      "language_loss": 0.828668,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85039073,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.6103403568267822
    },
    {
      "auxiliary_loss_clip": 0.01156672,
      "auxiliary_loss_mlp": 0.01028341,
      "balance_loss_clip": 1.04529738,
      "balance_loss_mlp": 1.02025902,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.5014774561432076,
      "language_loss": 0.74602699,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.7678771,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.590118408203125
    },
    {
      "auxiliary_loss_clip": 0.01068828,
      "auxiliary_loss_mlp": 0.0089161,
      "balance_loss_clip": 1.03567052,
      "balance_loss_mlp": 1.00015211,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8926463372311323,
      "language_loss": 0.6581049,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67770934,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.222062110900879
    },
    {
      "auxiliary_loss_clip": 0.0115986,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.04817986,
      "balance_loss_mlp": 1.01824331,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 2.1187440602591567,
      "language_loss": 0.73943263,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.76128334,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.7094264030456543
    },
    {
      "auxiliary_loss_clip": 0.01170548,
      "auxiliary_loss_mlp": 0.0102286,
      "balance_loss_clip": 1.04738963,
      "balance_loss_mlp": 1.01523685,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 2.0264154520732167,
      "language_loss": 0.84357154,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.8655057,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.603668451309204
    },
    {
      "auxiliary_loss_clip": 0.01059243,
      "auxiliary_loss_mlp": 0.01000794,
      "balance_loss_clip": 1.01139355,
      "balance_loss_mlp": 0.99988765,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7709574556919705,
      "language_loss": 0.58987272,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61047304,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2181057929992676
    },
    {
      "auxiliary_loss_clip": 0.01172515,
      "auxiliary_loss_mlp": 0.01027556,
      "balance_loss_clip": 1.04791975,
      "balance_loss_mlp": 1.01980472,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 1.8056378151145742,
      "language_loss": 0.74070132,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76270205,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 3.495537519454956
    },
    {
      "auxiliary_loss_clip": 0.01138305,
      "auxiliary_loss_mlp": 0.01027044,
      "balance_loss_clip": 1.04360557,
      "balance_loss_mlp": 1.01996875,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 1.8251922615151117,
      "language_loss": 0.75219417,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.7738477,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.9152448177337646
    },
    {
      "auxiliary_loss_clip": 0.01143451,
      "auxiliary_loss_mlp": 0.010265,
      "balance_loss_clip": 1.04691076,
      "balance_loss_mlp": 1.01915383,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.9373439062329414,
      "language_loss": 0.87908602,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.90078557,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.7141170501708984
    },
    {
      "auxiliary_loss_clip": 0.01141941,
      "auxiliary_loss_mlp": 0.00901789,
      "balance_loss_clip": 1.04424691,
      "balance_loss_mlp": 1.00063562,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 2.320942529308687,
      "language_loss": 0.71449268,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.73492998,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.879533290863037
    },
    {
      "auxiliary_loss_clip": 0.01141831,
      "auxiliary_loss_mlp": 0.01029068,
      "balance_loss_clip": 1.04743075,
      "balance_loss_mlp": 1.02184117,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.6140190804181425,
      "language_loss": 0.69775259,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71946156,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 3.785923719406128
    },
    {
      "auxiliary_loss_clip": 0.01179801,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 1.05384803,
      "balance_loss_mlp": 1.02336836,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 2.335901538715101,
      "language_loss": 0.78713715,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80924654,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.6069834232330322
    },
    {
      "auxiliary_loss_clip": 0.01160729,
      "auxiliary_loss_mlp": 0.01023289,
      "balance_loss_clip": 1.05090261,
      "balance_loss_mlp": 1.01637506,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.07694841519266,
      "language_loss": 0.78260207,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.80444229,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.5705482959747314
    },
    {
      "auxiliary_loss_clip": 0.01161303,
      "auxiliary_loss_mlp": 0.01022759,
      "balance_loss_clip": 1.04712677,
      "balance_loss_mlp": 1.01566029,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 2.8852525140893723,
      "language_loss": 0.8761397,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89798033,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.6501474380493164
    },
    {
      "auxiliary_loss_clip": 0.01176038,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.04906392,
      "balance_loss_mlp": 1.01871455,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 2.0568884455397485,
      "language_loss": 0.64115691,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.66318554,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.538649082183838
    },
    {
      "auxiliary_loss_clip": 0.01152517,
      "auxiliary_loss_mlp": 0.01026711,
      "balance_loss_clip": 1.04729462,
      "balance_loss_mlp": 1.01875341,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.9443943808427833,
      "language_loss": 0.84418893,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86598122,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 3.6559362411499023
    },
    {
      "auxiliary_loss_clip": 0.01147224,
      "auxiliary_loss_mlp": 0.00900695,
      "balance_loss_clip": 1.04391742,
      "balance_loss_mlp": 1.00061536,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 3.0267480970998815,
      "language_loss": 0.69725424,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.71773344,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.6601550579071045
    },
    {
      "auxiliary_loss_clip": 0.01163678,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.04974806,
      "balance_loss_mlp": 1.0182656,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.8499839741924835,
      "language_loss": 0.7585876,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78048635,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.568854331970215
    },
    {
      "auxiliary_loss_clip": 0.01170279,
      "auxiliary_loss_mlp": 0.01028565,
      "balance_loss_clip": 1.04833364,
      "balance_loss_mlp": 1.02089405,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.80852567412229,
      "language_loss": 0.74700975,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76899821,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.58359432220459
    },
    {
      "auxiliary_loss_clip": 0.01170194,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.04766631,
      "balance_loss_mlp": 1.01932299,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 2.117455698980217,
      "language_loss": 0.71705419,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.7390281,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.598804473876953
    },
    {
      "auxiliary_loss_clip": 0.01131809,
      "auxiliary_loss_mlp": 0.01022956,
      "balance_loss_clip": 1.0431664,
      "balance_loss_mlp": 1.01586616,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 2.1364369895064437,
      "language_loss": 0.7572996,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77884728,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.7984747886657715
    },
    {
      "auxiliary_loss_clip": 0.01159398,
      "auxiliary_loss_mlp": 0.01027044,
      "balance_loss_clip": 1.04768705,
      "balance_loss_mlp": 1.01951027,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 2.015882536267746,
      "language_loss": 0.82426,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84612441,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.582103729248047
    },
    {
      "auxiliary_loss_clip": 0.01169414,
      "auxiliary_loss_mlp": 0.01022202,
      "balance_loss_clip": 1.04707241,
      "balance_loss_mlp": 1.01470983,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 1.783021049738539,
      "language_loss": 0.65046966,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.67238581,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.6167068481445312
    },
    {
      "auxiliary_loss_clip": 0.01170909,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04903603,
      "balance_loss_mlp": 1.01715469,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 2.04102668410772,
      "language_loss": 0.75316274,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77511603,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.5664479732513428
    },
    {
      "auxiliary_loss_clip": 0.01117642,
      "auxiliary_loss_mlp": 0.01027282,
      "balance_loss_clip": 1.04218602,
      "balance_loss_mlp": 1.02022219,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 1.7947396979028551,
      "language_loss": 0.786883,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80833232,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 3.714106321334839
    },
    {
      "auxiliary_loss_clip": 0.0113562,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.04267383,
      "balance_loss_mlp": 1.01714706,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 1.8349281990433315,
      "language_loss": 0.65220177,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.6738019,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 3.01293683052063
    },
    {
      "auxiliary_loss_clip": 0.01155351,
      "auxiliary_loss_mlp": 0.01026642,
      "balance_loss_clip": 1.04536283,
      "balance_loss_mlp": 1.01867032,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.2597570375959677,
      "language_loss": 0.79583681,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.81765676,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.678523540496826
    },
    {
      "auxiliary_loss_clip": 0.01162033,
      "auxiliary_loss_mlp": 0.01029046,
      "balance_loss_clip": 1.04793131,
      "balance_loss_mlp": 1.02192378,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 2.342160003441082,
      "language_loss": 0.80993187,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.83184266,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.6709842681884766
    },
    {
      "auxiliary_loss_clip": 0.01171667,
      "auxiliary_loss_mlp": 0.01032096,
      "balance_loss_clip": 1.04882097,
      "balance_loss_mlp": 1.02428174,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 3.1083811604934075,
      "language_loss": 0.73067605,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75271368,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.6089041233062744
    },
    {
      "auxiliary_loss_clip": 0.01138189,
      "auxiliary_loss_mlp": 0.01022896,
      "balance_loss_clip": 1.04631388,
      "balance_loss_mlp": 1.01522493,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 3.7686013211891383,
      "language_loss": 0.76474714,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.786358,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.861992359161377
    },
    {
      "auxiliary_loss_clip": 0.0115617,
      "auxiliary_loss_mlp": 0.01023518,
      "balance_loss_clip": 1.04517508,
      "balance_loss_mlp": 1.01622868,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.7388527905618079,
      "language_loss": 0.84491599,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86671281,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.6728317737579346
    },
    {
      "auxiliary_loss_clip": 0.01164521,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.04728734,
      "balance_loss_mlp": 1.01851308,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.7487730952098381,
      "language_loss": 0.80772007,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.82962418,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.7078030109405518
    },
    {
      "auxiliary_loss_clip": 0.01068593,
      "auxiliary_loss_mlp": 0.01001164,
      "balance_loss_clip": 1.01179647,
      "balance_loss_mlp": 1.00029373,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7699696259419447,
      "language_loss": 0.61319679,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63389432,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.2788069248199463
    },
    {
      "auxiliary_loss_clip": 0.01163041,
      "auxiliary_loss_mlp": 0.01020262,
      "balance_loss_clip": 1.04634356,
      "balance_loss_mlp": 1.01277554,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.8979398498307138,
      "language_loss": 0.64216465,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66399765,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 3.5583336353302
    },
    {
      "auxiliary_loss_clip": 0.01150275,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.05330586,
      "balance_loss_mlp": 1.02230549,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 2.0733994895576098,
      "language_loss": 0.72145724,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74326277,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.803392171859741
    },
    {
      "auxiliary_loss_clip": 0.01152888,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.04607618,
      "balance_loss_mlp": 1.02143931,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.5501728729240893,
      "language_loss": 0.79840469,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.82022995,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.7072765827178955
    },
    {
      "auxiliary_loss_clip": 0.0115345,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 1.04712784,
      "balance_loss_mlp": 1.0229727,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 2.0369813902660456,
      "language_loss": 0.79421729,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81605905,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.7497189044952393
    },
    {
      "auxiliary_loss_clip": 0.01154194,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.05095983,
      "balance_loss_mlp": 1.02033627,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.21520600765007,
      "language_loss": 0.68869913,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.71051645,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 3.7922091484069824
    },
    {
      "auxiliary_loss_clip": 0.01159274,
      "auxiliary_loss_mlp": 0.01020959,
      "balance_loss_clip": 1.04513001,
      "balance_loss_mlp": 1.0133121,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.0549006857779966,
      "language_loss": 0.73562568,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75742805,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.6098392009735107
    },
    {
      "auxiliary_loss_clip": 0.01141076,
      "auxiliary_loss_mlp": 0.01029866,
      "balance_loss_clip": 1.04169703,
      "balance_loss_mlp": 1.02103817,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 2.1904620883112775,
      "language_loss": 0.62726891,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.64897835,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 2.983994483947754
    },
    {
      "auxiliary_loss_clip": 0.01162537,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.04910421,
      "balance_loss_mlp": 1.01840067,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.924835435666447,
      "language_loss": 0.77232856,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79421484,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.7508065700531006
    },
    {
      "auxiliary_loss_clip": 0.01172727,
      "auxiliary_loss_mlp": 0.01026079,
      "balance_loss_clip": 1.04988122,
      "balance_loss_mlp": 1.01849782,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.8425542554879,
      "language_loss": 0.76401514,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78600323,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.622039794921875
    },
    {
      "auxiliary_loss_clip": 0.0115658,
      "auxiliary_loss_mlp": 0.01023402,
      "balance_loss_clip": 1.04928148,
      "balance_loss_mlp": 1.0159868,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.9636829392836568,
      "language_loss": 0.72762811,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74942797,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 3.647251844406128
    },
    {
      "auxiliary_loss_clip": 0.01150521,
      "auxiliary_loss_mlp": 0.00901028,
      "balance_loss_clip": 1.04534233,
      "balance_loss_mlp": 1.00067735,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 1.8496973956085234,
      "language_loss": 0.76752102,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78803653,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.70797061920166
    },
    {
      "auxiliary_loss_clip": 0.01142122,
      "auxiliary_loss_mlp": 0.01022922,
      "balance_loss_clip": 1.04314709,
      "balance_loss_mlp": 1.01586485,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.5739856337390234,
      "language_loss": 0.75566065,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77731115,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.7553436756134033
    },
    {
      "auxiliary_loss_clip": 0.0115201,
      "auxiliary_loss_mlp": 0.01027606,
      "balance_loss_clip": 1.04504204,
      "balance_loss_mlp": 1.01929152,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 2.811321435554501,
      "language_loss": 0.60274744,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62454367,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.6451845169067383
    },
    {
      "auxiliary_loss_clip": 0.01156593,
      "auxiliary_loss_mlp": 0.01029912,
      "balance_loss_clip": 1.04730654,
      "balance_loss_mlp": 1.02246785,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 3.8452310612521665,
      "language_loss": 0.72656459,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.74842966,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.6455447673797607
    },
    {
      "auxiliary_loss_clip": 0.01126221,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.04114532,
      "balance_loss_mlp": 1.01927471,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.4019287708416233,
      "language_loss": 0.79513752,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81667101,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.7328741550445557
    },
    {
      "auxiliary_loss_clip": 0.01165069,
      "auxiliary_loss_mlp": 0.01025164,
      "balance_loss_clip": 1.04725814,
      "balance_loss_mlp": 1.01768088,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.6668025257653976,
      "language_loss": 0.75415516,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77605742,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.702238082885742
    },
    {
      "auxiliary_loss_clip": 0.01157537,
      "auxiliary_loss_mlp": 0.01023692,
      "balance_loss_clip": 1.0466361,
      "balance_loss_mlp": 1.01610684,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 2.969447187008325,
      "language_loss": 0.76719421,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78900653,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.6054699420928955
    },
    {
      "auxiliary_loss_clip": 0.01127729,
      "auxiliary_loss_mlp": 0.01027701,
      "balance_loss_clip": 1.04406655,
      "balance_loss_mlp": 1.01984775,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 2.795964949041253,
      "language_loss": 0.79208541,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81363964,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 3.858144998550415
    },
    {
      "auxiliary_loss_clip": 0.01132523,
      "auxiliary_loss_mlp": 0.0102471,
      "balance_loss_clip": 1.04195833,
      "balance_loss_mlp": 1.01713443,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 1.974369551816081,
      "language_loss": 0.76624525,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.7878176,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.774502992630005
    },
    {
      "auxiliary_loss_clip": 0.01166865,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.04925871,
      "balance_loss_mlp": 1.01655221,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 2.351001358069436,
      "language_loss": 0.69785881,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.71977508,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.6188809871673584
    },
    {
      "auxiliary_loss_clip": 0.01174209,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 1.05113947,
      "balance_loss_mlp": 1.01707578,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.0633542790959076,
      "language_loss": 0.78392112,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80591154,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.6055779457092285
    },
    {
      "auxiliary_loss_clip": 0.01052178,
      "auxiliary_loss_mlp": 0.01002014,
      "balance_loss_clip": 1.01164615,
      "balance_loss_mlp": 1.00117397,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7932639724523538,
      "language_loss": 0.53946745,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.56000936,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.3049378395080566
    },
    {
      "auxiliary_loss_clip": 0.01152232,
      "auxiliary_loss_mlp": 0.01034538,
      "balance_loss_clip": 1.04625356,
      "balance_loss_mlp": 1.02673841,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.2155002943018562,
      "language_loss": 0.68616623,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70803392,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.818552255630493
    },
    {
      "auxiliary_loss_clip": 0.01152812,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.04690886,
      "balance_loss_mlp": 1.0187006,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.1629549067663763,
      "language_loss": 0.68472028,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70652431,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.766775131225586
    },
    {
      "auxiliary_loss_clip": 0.01173817,
      "auxiliary_loss_mlp": 0.01027245,
      "balance_loss_clip": 1.05018651,
      "balance_loss_mlp": 1.01971436,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.3366496659689595,
      "language_loss": 0.72781754,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74982816,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.6528234481811523
    },
    {
      "auxiliary_loss_clip": 0.01154181,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.04760277,
      "balance_loss_mlp": 1.0162698,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.069613082355404,
      "language_loss": 0.69891357,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72069186,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.6664907932281494
    },
    {
      "auxiliary_loss_clip": 0.01149369,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.04250455,
      "balance_loss_mlp": 1.01878357,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 2.3438106540292276,
      "language_loss": 0.75542188,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77717918,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 3.6050350666046143
    },
    {
      "auxiliary_loss_clip": 0.01158881,
      "auxiliary_loss_mlp": 0.00900621,
      "balance_loss_clip": 1.04886484,
      "balance_loss_mlp": 1.00060987,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.381128475111804,
      "language_loss": 0.77981019,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.80040514,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.6720848083496094
    },
    {
      "auxiliary_loss_clip": 0.01165091,
      "auxiliary_loss_mlp": 0.01029388,
      "balance_loss_clip": 1.05037248,
      "balance_loss_mlp": 1.0221225,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.6577917607981034,
      "language_loss": 0.70850778,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73045254,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.6304986476898193
    },
    {
      "auxiliary_loss_clip": 0.01130415,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.0392853,
      "balance_loss_mlp": 1.01822662,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 2.010791648725373,
      "language_loss": 0.86088395,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.882447,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.7277615070343018
    },
    {
      "auxiliary_loss_clip": 0.01165389,
      "auxiliary_loss_mlp": 0.00901234,
      "balance_loss_clip": 1.04550683,
      "balance_loss_mlp": 1.00065255,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 1.9607429825468214,
      "language_loss": 0.81151938,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83218563,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 3.639228343963623
    },
    {
      "auxiliary_loss_clip": 0.01173903,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.05022657,
      "balance_loss_mlp": 1.01827502,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 2.0655231480960343,
      "language_loss": 0.8486771,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87067854,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.549926280975342
    },
    {
      "auxiliary_loss_clip": 0.01147419,
      "auxiliary_loss_mlp": 0.01021445,
      "balance_loss_clip": 1.04947948,
      "balance_loss_mlp": 1.01431298,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 12.568734591260142,
      "language_loss": 0.68363094,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70531964,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.7931370735168457
    },
    {
      "auxiliary_loss_clip": 0.0115303,
      "auxiliary_loss_mlp": 0.01026048,
      "balance_loss_clip": 1.04644012,
      "balance_loss_mlp": 1.01862442,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 1.743043233530368,
      "language_loss": 0.7112416,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73303235,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.6127512454986572
    },
    {
      "auxiliary_loss_clip": 0.01165362,
      "auxiliary_loss_mlp": 0.01023698,
      "balance_loss_clip": 1.04958665,
      "balance_loss_mlp": 1.01659274,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.9288251200455706,
      "language_loss": 0.86045092,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.8823415,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.6881048679351807
    },
    {
      "auxiliary_loss_clip": 0.01149069,
      "auxiliary_loss_mlp": 0.01031031,
      "balance_loss_clip": 1.04607797,
      "balance_loss_mlp": 1.02372313,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 1.7849890562850557,
      "language_loss": 0.73226929,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75407034,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.6937477588653564
    },
    {
      "auxiliary_loss_clip": 0.0115837,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04515767,
      "balance_loss_mlp": 1.0194118,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.7784967237933238,
      "language_loss": 0.79131794,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.81316757,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 3.5755977630615234
    },
    {
      "auxiliary_loss_clip": 0.01154886,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.04722261,
      "balance_loss_mlp": 1.01944745,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 2.8900246348178484,
      "language_loss": 0.64945281,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67127979,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.7207024097442627
    },
    {
      "auxiliary_loss_clip": 0.01154586,
      "auxiliary_loss_mlp": 0.01021382,
      "balance_loss_clip": 1.048841,
      "balance_loss_mlp": 1.01383352,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 1.9567021734198422,
      "language_loss": 0.71832943,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.74008918,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.625823497772217
    },
    {
      "auxiliary_loss_clip": 0.01155087,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.04815221,
      "balance_loss_mlp": 1.02073574,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.390387654407813,
      "language_loss": 0.80406147,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.82589984,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.6332521438598633
    },
    {
      "auxiliary_loss_clip": 0.01068918,
      "auxiliary_loss_mlp": 0.01000807,
      "balance_loss_clip": 1.01268077,
      "balance_loss_mlp": 1.00000215,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6463484013573496,
      "language_loss": 0.60312253,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62381983,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.2474913597106934
    },
    {
      "auxiliary_loss_clip": 0.01140922,
      "auxiliary_loss_mlp": 0.01025405,
      "balance_loss_clip": 1.04189551,
      "balance_loss_mlp": 1.01809192,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.6894403727993768,
      "language_loss": 0.83046657,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85212988,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.738128185272217
    },
    {
      "auxiliary_loss_clip": 0.01136888,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.04330719,
      "balance_loss_mlp": 1.01580215,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 9.855224151789056,
      "language_loss": 0.80037338,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82197875,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.6422924995422363
    },
    {
      "auxiliary_loss_clip": 0.01175152,
      "auxiliary_loss_mlp": 0.01024735,
      "balance_loss_clip": 1.04869449,
      "balance_loss_mlp": 1.01691747,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 2.5277896555169135,
      "language_loss": 0.62732714,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64932603,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.6645843982696533
    },
    {
      "auxiliary_loss_clip": 0.01173688,
      "auxiliary_loss_mlp": 0.0102445,
      "balance_loss_clip": 1.05022645,
      "balance_loss_mlp": 1.01682699,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.373828727939661,
      "language_loss": 0.8532207,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87520206,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 3.5183069705963135
    },
    {
      "auxiliary_loss_clip": 0.01159034,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.0532968,
      "balance_loss_mlp": 1.02146077,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.4821419646809466,
      "language_loss": 0.63796312,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.65984499,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.7586846351623535
    },
    {
      "auxiliary_loss_clip": 0.01154641,
      "auxiliary_loss_mlp": 0.01025252,
      "balance_loss_clip": 1.04588413,
      "balance_loss_mlp": 1.01724744,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.1722866719747613,
      "language_loss": 0.72250223,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.7443012,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.627362012863159
    },
    {
      "auxiliary_loss_clip": 0.01164496,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.04982448,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.741427501023017,
      "language_loss": 0.81046093,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83237368,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.6109774112701416
    },
    {
      "auxiliary_loss_clip": 0.01154616,
      "auxiliary_loss_mlp": 0.01028192,
      "balance_loss_clip": 1.045681,
      "balance_loss_mlp": 1.02095032,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 5.333920519079503,
      "language_loss": 0.79572755,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81755567,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.6059563159942627
    },
    {
      "auxiliary_loss_clip": 0.01170126,
      "auxiliary_loss_mlp": 0.0102384,
      "balance_loss_clip": 1.04859161,
      "balance_loss_mlp": 1.01695836,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 1.923722150450648,
      "language_loss": 0.7939418,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81588149,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.5109386444091797
    },
    {
      "auxiliary_loss_clip": 0.01173197,
      "auxiliary_loss_mlp": 0.01026311,
      "balance_loss_clip": 1.04826558,
      "balance_loss_mlp": 1.01828814,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.595370918314747,
      "language_loss": 0.75273919,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77473426,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.5935232639312744
    },
    {
      "auxiliary_loss_clip": 0.01163803,
      "auxiliary_loss_mlp": 0.01023501,
      "balance_loss_clip": 1.04788148,
      "balance_loss_mlp": 1.01574647,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.9286467547084827,
      "language_loss": 0.8328253,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.8546983,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.640326738357544
    },
    {
      "auxiliary_loss_clip": 0.01072422,
      "auxiliary_loss_mlp": 0.01000981,
      "balance_loss_clip": 1.01223016,
      "balance_loss_mlp": 1.00018215,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.8083635057515743,
      "language_loss": 0.58967835,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61041236,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 4.087154865264893
    },
    {
      "auxiliary_loss_clip": 0.01144905,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.04399872,
      "balance_loss_mlp": 1.01847601,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.1630473399551797,
      "language_loss": 0.66393459,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68564701,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.739440441131592
    },
    {
      "auxiliary_loss_clip": 0.01162923,
      "auxiliary_loss_mlp": 0.00901074,
      "balance_loss_clip": 1.05361807,
      "balance_loss_mlp": 1.00069976,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.757633116013262,
      "language_loss": 0.70099348,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72163343,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.6489689350128174
    },
    {
      "auxiliary_loss_clip": 0.01168031,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.04874349,
      "balance_loss_mlp": 1.01800311,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.6052933267374594,
      "language_loss": 0.70962024,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73155677,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.825697183609009
    },
    {
      "auxiliary_loss_clip": 0.01149005,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 1.04677606,
      "balance_loss_mlp": 1.01760483,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.5616281408927506,
      "language_loss": 0.73408949,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75583255,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.648005485534668
    },
    {
      "auxiliary_loss_clip": 0.01142227,
      "auxiliary_loss_mlp": 0.01023513,
      "balance_loss_clip": 1.04238081,
      "balance_loss_mlp": 1.01577652,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 3.0166377537282645,
      "language_loss": 0.61885363,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.64051104,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 3.694411039352417
    },
    {
      "auxiliary_loss_clip": 0.01148913,
      "auxiliary_loss_mlp": 0.01024785,
      "balance_loss_clip": 1.04778051,
      "balance_loss_mlp": 1.01757574,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.381841574763479,
      "language_loss": 0.75307095,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77480793,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.6463966369628906
    },
    {
      "auxiliary_loss_clip": 0.01161654,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.04752779,
      "balance_loss_mlp": 1.01560462,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.9994514624024033,
      "language_loss": 0.75273848,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77458608,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.57578706741333
    },
    {
      "auxiliary_loss_clip": 0.01162686,
      "auxiliary_loss_mlp": 0.01025109,
      "balance_loss_clip": 1.04905021,
      "balance_loss_mlp": 1.01752114,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.8755850236190428,
      "language_loss": 0.70203954,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.72391748,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.600728750228882
    },
    {
      "auxiliary_loss_clip": 0.01161151,
      "auxiliary_loss_mlp": 0.01021425,
      "balance_loss_clip": 1.04536963,
      "balance_loss_mlp": 1.01388526,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 1.6909768222684216,
      "language_loss": 0.77135026,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79317605,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.683228015899658
    },
    {
      "auxiliary_loss_clip": 0.01166109,
      "auxiliary_loss_mlp": 0.01023079,
      "balance_loss_clip": 1.04658866,
      "balance_loss_mlp": 1.01569998,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 2.5306156031644282,
      "language_loss": 0.72154522,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.74343705,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 3.616010904312134
    },
    {
      "auxiliary_loss_clip": 0.0116448,
      "auxiliary_loss_mlp": 0.01022319,
      "balance_loss_clip": 1.04782677,
      "balance_loss_mlp": 1.01488912,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.3179097070524888,
      "language_loss": 0.7391187,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76098669,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.6552460193634033
    },
    {
      "auxiliary_loss_clip": 0.01160515,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.04357219,
      "balance_loss_mlp": 1.02143216,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.5165238479480236,
      "language_loss": 0.68212634,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70402098,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.8824377059936523
    },
    {
      "auxiliary_loss_clip": 0.0116827,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.04787588,
      "balance_loss_mlp": 1.02105689,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.5945887870960163,
      "language_loss": 0.64005947,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.6620298,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.6436939239501953
    },
    {
      "auxiliary_loss_clip": 0.01170013,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.04802608,
      "balance_loss_mlp": 1.01880455,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 3.896398306418406,
      "language_loss": 0.76828206,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.7902506,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.542400598526001
    },
    {
      "auxiliary_loss_clip": 0.01047773,
      "auxiliary_loss_mlp": 0.01006655,
      "balance_loss_clip": 1.01316428,
      "balance_loss_mlp": 1.00562382,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7387615111008026,
      "language_loss": 0.58299625,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60354054,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.1131715774536133
    },
    {
      "auxiliary_loss_clip": 0.01068676,
      "auxiliary_loss_mlp": 0.01001714,
      "balance_loss_clip": 1.01301241,
      "balance_loss_mlp": 1.00084364,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8438901501184063,
      "language_loss": 0.56909561,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58979952,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.277256488800049
    },
    {
      "auxiliary_loss_clip": 0.01065039,
      "auxiliary_loss_mlp": 0.01000924,
      "balance_loss_clip": 1.01409245,
      "balance_loss_mlp": 1.00012517,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8720827514379226,
      "language_loss": 0.67442,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69507968,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.0770628452301025
    },
    {
      "auxiliary_loss_clip": 0.01138421,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.05015099,
      "balance_loss_mlp": 1.0170939,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.7035246633187628,
      "language_loss": 0.76269716,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78432548,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 3.6609959602355957
    },
    {
      "auxiliary_loss_clip": 0.01154097,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.04669237,
      "balance_loss_mlp": 1.02149892,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.7616427932147674,
      "language_loss": 0.83156347,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85340238,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.63980770111084
    },
    {
      "auxiliary_loss_clip": 0.01167147,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.04861951,
      "balance_loss_mlp": 1.0204041,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.4920175119805732,
      "language_loss": 0.76061893,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78256059,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.6828861236572266
    },
    {
      "auxiliary_loss_clip": 0.0113816,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04259217,
      "balance_loss_mlp": 1.01841497,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.786729241536207,
      "language_loss": 0.86507821,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88670927,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.7385776042938232
    },
    {
      "auxiliary_loss_clip": 0.01172447,
      "auxiliary_loss_mlp": 0.01022999,
      "balance_loss_clip": 1.04889965,
      "balance_loss_mlp": 1.01545882,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 1.7747326418888725,
      "language_loss": 0.75983179,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78178626,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.5808210372924805
    },
    {
      "auxiliary_loss_clip": 0.01161826,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.04540133,
      "balance_loss_mlp": 1.01615429,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.6935185550064564,
      "language_loss": 0.8049953,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82685006,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.5935325622558594
    },
    {
      "auxiliary_loss_clip": 0.01161114,
      "auxiliary_loss_mlp": 0.01024222,
      "balance_loss_clip": 1.04623866,
      "balance_loss_mlp": 1.01697099,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.7642311840336697,
      "language_loss": 0.76844144,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.79029477,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.586740493774414
    },
    {
      "auxiliary_loss_clip": 0.01155382,
      "auxiliary_loss_mlp": 0.01026985,
      "balance_loss_clip": 1.04456389,
      "balance_loss_mlp": 1.01975822,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.702083739738406,
      "language_loss": 0.77506196,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79688561,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.6198158264160156
    },
    {
      "auxiliary_loss_clip": 0.0114897,
      "auxiliary_loss_mlp": 0.01024936,
      "balance_loss_clip": 1.04473782,
      "balance_loss_mlp": 1.01733065,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 1.8333307660598006,
      "language_loss": 0.78570199,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80744106,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.6620659828186035
    },
    {
      "auxiliary_loss_clip": 0.01166773,
      "auxiliary_loss_mlp": 0.00901172,
      "balance_loss_clip": 1.04767466,
      "balance_loss_mlp": 1.00079286,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.7234357440691646,
      "language_loss": 0.8336342,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.85431361,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 3.445021629333496
    },
    {
      "auxiliary_loss_clip": 0.01164037,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.04697204,
      "balance_loss_mlp": 1.02103281,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.6339701982992927,
      "language_loss": 0.81925422,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84117627,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.615046739578247
    },
    {
      "auxiliary_loss_clip": 0.01071941,
      "auxiliary_loss_mlp": 0.01001113,
      "balance_loss_clip": 1.01201534,
      "balance_loss_mlp": 1.00033808,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6860113489727988,
      "language_loss": 0.55946171,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58019233,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.153839111328125
    },
    {
      "auxiliary_loss_clip": 0.01166969,
      "auxiliary_loss_mlp": 0.01027666,
      "balance_loss_clip": 1.05283415,
      "balance_loss_mlp": 1.01961398,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.742965753444415,
      "language_loss": 0.76615322,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.78809965,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.867276668548584
    },
    {
      "auxiliary_loss_clip": 0.01157552,
      "auxiliary_loss_mlp": 0.01023259,
      "balance_loss_clip": 1.04907882,
      "balance_loss_mlp": 1.01531386,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 3.519556821041532,
      "language_loss": 0.6728313,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69463938,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 3.5708680152893066
    },
    {
      "auxiliary_loss_clip": 0.01153099,
      "auxiliary_loss_mlp": 0.0102766,
      "balance_loss_clip": 1.04473591,
      "balance_loss_mlp": 1.02032268,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 2.0092964132576823,
      "language_loss": 0.7725845,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.79439205,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.748506784439087
    },
    {
      "auxiliary_loss_clip": 0.01167878,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.05167198,
      "balance_loss_mlp": 1.01714492,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 1.8989740099015506,
      "language_loss": 0.77998662,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.80191183,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.6147730350494385
    },
    {
      "auxiliary_loss_clip": 0.01158123,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.05129564,
      "balance_loss_mlp": 1.01913738,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.64254173167812,
      "language_loss": 0.79458344,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81642687,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.632066249847412
    },
    {
      "auxiliary_loss_clip": 0.01170518,
      "auxiliary_loss_mlp": 0.01023716,
      "balance_loss_clip": 1.04906559,
      "balance_loss_mlp": 1.01683426,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.9238316736907788,
      "language_loss": 0.74085432,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76279664,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.5330870151519775
    },
    {
      "auxiliary_loss_clip": 0.01132864,
      "auxiliary_loss_mlp": 0.00901467,
      "balance_loss_clip": 1.0411787,
      "balance_loss_mlp": 1.0007571,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.8301640593293553,
      "language_loss": 0.74539959,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.7657429,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 3.6794118881225586
    },
    {
      "auxiliary_loss_clip": 0.0115462,
      "auxiliary_loss_mlp": 0.01028902,
      "balance_loss_clip": 1.05237818,
      "balance_loss_mlp": 1.02166593,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.9150123965917025,
      "language_loss": 0.72579086,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74762613,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.5729966163635254
    },
    {
      "auxiliary_loss_clip": 0.01169711,
      "auxiliary_loss_mlp": 0.01020486,
      "balance_loss_clip": 1.04680371,
      "balance_loss_mlp": 1.01346159,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.3249611365739447,
      "language_loss": 0.72526956,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74717152,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.6046011447906494
    },
    {
      "auxiliary_loss_clip": 0.01155528,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.04494262,
      "balance_loss_mlp": 1.01816308,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.3001104338475584,
      "language_loss": 0.84186667,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.8636874,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.6098849773406982
    },
    {
      "auxiliary_loss_clip": 0.01152328,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 1.04574609,
      "balance_loss_mlp": 1.01792896,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 2.048232506763382,
      "language_loss": 0.66895223,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.69072896,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.867687702178955
    },
    {
      "auxiliary_loss_clip": 0.0105494,
      "auxiliary_loss_mlp": 0.01002501,
      "balance_loss_clip": 1.01218438,
      "balance_loss_mlp": 1.00152934,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8421759264407146,
      "language_loss": 0.52425086,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54482532,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.1796326637268066
    },
    {
      "auxiliary_loss_clip": 0.01139836,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 1.04442871,
      "balance_loss_mlp": 1.01657546,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.8680290573769573,
      "language_loss": 0.73129451,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75293791,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.6330552101135254
    },
    {
      "auxiliary_loss_clip": 0.01150272,
      "auxiliary_loss_mlp": 0.01020648,
      "balance_loss_clip": 1.04558587,
      "balance_loss_mlp": 1.0131799,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.234811970617585,
      "language_loss": 0.71931142,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74102068,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.6736791133880615
    },
    {
      "auxiliary_loss_clip": 0.01145846,
      "auxiliary_loss_mlp": 0.01022004,
      "balance_loss_clip": 1.04463768,
      "balance_loss_mlp": 1.01484871,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.4395694613871135,
      "language_loss": 0.73323292,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75491142,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.6488304138183594
    },
    {
      "auxiliary_loss_clip": 0.01165385,
      "auxiliary_loss_mlp": 0.0102736,
      "balance_loss_clip": 1.04581928,
      "balance_loss_mlp": 1.01975441,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.865795715641123,
      "language_loss": 0.78029954,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80222696,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 3.566054582595825
    },
    {
      "auxiliary_loss_clip": 0.01137001,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.03966177,
      "balance_loss_mlp": 1.01927722,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.6246460710238055,
      "language_loss": 0.74950302,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.77113783,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.6647746562957764
    },
    {
      "auxiliary_loss_clip": 0.01159271,
      "auxiliary_loss_mlp": 0.00901671,
      "balance_loss_clip": 1.04595304,
      "balance_loss_mlp": 1.00067186,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.9027021217721443,
      "language_loss": 0.65958554,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.68019497,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.8364157676696777
    },
    {
      "auxiliary_loss_clip": 0.01072017,
      "auxiliary_loss_mlp": 0.00890814,
      "balance_loss_clip": 1.01251173,
      "balance_loss_mlp": 0.99996108,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7050784008694778,
      "language_loss": 0.55724794,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57687634,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.0936524868011475
    },
    {
      "auxiliary_loss_clip": 0.01165673,
      "auxiliary_loss_mlp": 0.01022792,
      "balance_loss_clip": 1.04484177,
      "balance_loss_mlp": 1.01473391,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.5830301481280307,
      "language_loss": 0.83140922,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.8532939,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.6273441314697266
    },
    {
      "auxiliary_loss_clip": 0.01174582,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.0499742,
      "balance_loss_mlp": 1.01969028,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 3.3190819290117766,
      "language_loss": 0.84222353,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.86424744,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.6190669536590576
    },
    {
      "auxiliary_loss_clip": 0.01161996,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.04717255,
      "balance_loss_mlp": 1.01789403,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.0581825299643315,
      "language_loss": 0.74281734,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76469094,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.592780590057373
    },
    {
      "auxiliary_loss_clip": 0.01150334,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 1.04660439,
      "balance_loss_mlp": 1.01963091,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.1144065359163835,
      "language_loss": 0.81427073,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83604693,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.7007384300231934
    },
    {
      "auxiliary_loss_clip": 0.01069002,
      "auxiliary_loss_mlp": 0.01000899,
      "balance_loss_clip": 1.01268017,
      "balance_loss_mlp": 1.00006413,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7140838816224745,
      "language_loss": 0.56316829,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58386731,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.1317074298858643
    },
    {
      "auxiliary_loss_clip": 0.01144152,
      "auxiliary_loss_mlp": 0.01025785,
      "balance_loss_clip": 1.0430845,
      "balance_loss_mlp": 1.01761317,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.2128579854510946,
      "language_loss": 0.74500322,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76670265,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 3.677605628967285
    },
    {
      "auxiliary_loss_clip": 0.01146621,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04591489,
      "balance_loss_mlp": 1.0174613,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.7572379262567825,
      "language_loss": 0.81982654,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84153509,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.738420248031616
    },
    {
      "auxiliary_loss_clip": 0.01162274,
      "auxiliary_loss_mlp": 0.01036627,
      "balance_loss_clip": 1.05025744,
      "balance_loss_mlp": 1.02872968,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 2.3319195979192164,
      "language_loss": 0.68045086,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.7024399,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.56977915763855
    },
    {
      "auxiliary_loss_clip": 0.01161118,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.05010498,
      "balance_loss_mlp": 1.01975918,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.207414338493138,
      "language_loss": 0.73252904,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.75441933,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 3.6370487213134766
    },
    {
      "auxiliary_loss_clip": 0.01120716,
      "auxiliary_loss_mlp": 0.01029923,
      "balance_loss_clip": 1.0386107,
      "balance_loss_mlp": 1.02240729,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.446530200465004,
      "language_loss": 0.8202467,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84175313,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.703026056289673
    },
    {
      "auxiliary_loss_clip": 0.01156315,
      "auxiliary_loss_mlp": 0.01023715,
      "balance_loss_clip": 1.0440706,
      "balance_loss_mlp": 1.01618063,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.8815463605019174,
      "language_loss": 0.77441072,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79621106,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.659055709838867
    },
    {
      "auxiliary_loss_clip": 0.01154929,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.0469923,
      "balance_loss_mlp": 1.01919436,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 2.479358837131347,
      "language_loss": 0.7910099,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.81282282,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.6304399967193604
    },
    {
      "auxiliary_loss_clip": 0.01147192,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.04562879,
      "balance_loss_mlp": 1.0190165,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 1.646859537856994,
      "language_loss": 0.8318249,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.853567,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.6594417095184326
    },
    {
      "auxiliary_loss_clip": 0.01155648,
      "auxiliary_loss_mlp": 0.01033438,
      "balance_loss_clip": 1.04479337,
      "balance_loss_mlp": 1.02596021,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.130123055620662,
      "language_loss": 0.80871761,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83060849,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.6606855392456055
    },
    {
      "auxiliary_loss_clip": 0.0115813,
      "auxiliary_loss_mlp": 0.01023617,
      "balance_loss_clip": 1.04685736,
      "balance_loss_mlp": 1.01607108,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 5.549633486714874,
      "language_loss": 0.76645404,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78827155,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 3.5920603275299072
    },
    {
      "auxiliary_loss_clip": 0.01162472,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.0492878,
      "balance_loss_mlp": 1.01828682,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.6896968367401446,
      "language_loss": 0.84391189,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86579382,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.6311779022216797
    },
    {
      "auxiliary_loss_clip": 0.01150425,
      "auxiliary_loss_mlp": 0.01028838,
      "balance_loss_clip": 1.04928315,
      "balance_loss_mlp": 1.02162325,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.9394013231757428,
      "language_loss": 0.75866538,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.78045803,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.683262825012207
    },
    {
      "auxiliary_loss_clip": 0.01141079,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.04456806,
      "balance_loss_mlp": 1.01919711,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.7101952516791146,
      "language_loss": 0.71073771,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.7324121,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.7554619312286377
    },
    {
      "auxiliary_loss_clip": 0.01149151,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.04782784,
      "balance_loss_mlp": 1.01969743,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 2.0938819826061863,
      "language_loss": 0.70515972,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72692204,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.6495447158813477
    },
    {
      "auxiliary_loss_clip": 0.01149411,
      "auxiliary_loss_mlp": 0.01028998,
      "balance_loss_clip": 1.04587269,
      "balance_loss_mlp": 1.02161336,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 2.1206167773476214,
      "language_loss": 0.68442547,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.7062096,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.6734468936920166
    },
    {
      "auxiliary_loss_clip": 0.01167743,
      "auxiliary_loss_mlp": 0.01031343,
      "balance_loss_clip": 1.04772449,
      "balance_loss_mlp": 1.0233798,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.83048948242193,
      "language_loss": 0.65152001,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67351091,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.6471660137176514
    },
    {
      "auxiliary_loss_clip": 0.01062206,
      "auxiliary_loss_mlp": 0.01000553,
      "balance_loss_clip": 1.01055384,
      "balance_loss_mlp": 0.99976635,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8025506884084068,
      "language_loss": 0.61335254,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63398015,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.14461350440979
    },
    {
      "auxiliary_loss_clip": 0.01174056,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.04883087,
      "balance_loss_mlp": 1.01905012,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.56545503419864,
      "language_loss": 0.83979034,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86180532,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 3.462695837020874
    },
    {
      "auxiliary_loss_clip": 0.01159783,
      "auxiliary_loss_mlp": 0.010248,
      "balance_loss_clip": 1.04537272,
      "balance_loss_mlp": 1.01773703,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.705375569696247,
      "language_loss": 0.8029961,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82484198,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.5961828231811523
    },
    {
      "auxiliary_loss_clip": 0.0106797,
      "auxiliary_loss_mlp": 0.01001017,
      "balance_loss_clip": 1.01348996,
      "balance_loss_mlp": 1.00023043,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.8110381859145656,
      "language_loss": 0.54815429,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.56884414,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.1765799522399902
    },
    {
      "auxiliary_loss_clip": 0.01165413,
      "auxiliary_loss_mlp": 0.01022696,
      "balance_loss_clip": 1.04765153,
      "balance_loss_mlp": 1.01524532,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 2.0885725877639367,
      "language_loss": 0.76947463,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79135573,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.543006658554077
    },
    {
      "auxiliary_loss_clip": 0.01138729,
      "auxiliary_loss_mlp": 0.01022759,
      "balance_loss_clip": 1.04231215,
      "balance_loss_mlp": 1.01490259,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.9525901281624143,
      "language_loss": 0.71866524,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.74028015,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.738460063934326
    },
    {
      "auxiliary_loss_clip": 0.01172236,
      "auxiliary_loss_mlp": 0.01021539,
      "balance_loss_clip": 1.05104446,
      "balance_loss_mlp": 1.01427293,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.9203453704216873,
      "language_loss": 0.87496006,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89689785,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.5881149768829346
    },
    {
      "auxiliary_loss_clip": 0.01142812,
      "auxiliary_loss_mlp": 0.01019365,
      "balance_loss_clip": 1.04567385,
      "balance_loss_mlp": 1.01234388,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.9504052798951859,
      "language_loss": 0.75164992,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77327174,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.696985960006714
    },
    {
      "auxiliary_loss_clip": 0.01062211,
      "auxiliary_loss_mlp": 0.01000873,
      "balance_loss_clip": 1.01047945,
      "balance_loss_mlp": 1.00006235,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8501269170086647,
      "language_loss": 0.60956627,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63019711,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.0934882164001465
    },
    {
      "auxiliary_loss_clip": 0.01164819,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.0464592,
      "balance_loss_mlp": 1.0192318,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.4794346698150336,
      "language_loss": 0.73232675,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.7542364,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.5587801933288574
    },
    {
      "auxiliary_loss_clip": 0.01138949,
      "auxiliary_loss_mlp": 0.01024894,
      "balance_loss_clip": 1.04191399,
      "balance_loss_mlp": 1.01680577,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.3410913518493293,
      "language_loss": 0.5955897,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.61722815,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 2.7596595287323
    },
    {
      "auxiliary_loss_clip": 0.01164627,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.05071688,
      "balance_loss_mlp": 1.01796889,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.3391163002379636,
      "language_loss": 0.88957417,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91147506,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 3.5261452198028564
    },
    {
      "auxiliary_loss_clip": 0.01164128,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04803348,
      "balance_loss_mlp": 1.01961577,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 2.4150559805900738,
      "language_loss": 0.8116563,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83356452,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.668635368347168
    },
    {
      "auxiliary_loss_clip": 0.01164791,
      "auxiliary_loss_mlp": 0.00901222,
      "balance_loss_clip": 1.05345726,
      "balance_loss_mlp": 1.00073504,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.636094942552297,
      "language_loss": 0.69240665,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71306676,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.648852825164795
    },
    {
      "auxiliary_loss_clip": 0.01146246,
      "auxiliary_loss_mlp": 0.01021276,
      "balance_loss_clip": 1.04621387,
      "balance_loss_mlp": 1.01476121,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.5511899693809272,
      "language_loss": 0.73723716,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75891232,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 3.5296409130096436
    },
    {
      "auxiliary_loss_clip": 0.01154336,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.04713178,
      "balance_loss_mlp": 1.01748562,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 2.2179523940847865,
      "language_loss": 0.7481513,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.76994205,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.6889476776123047
    },
    {
      "auxiliary_loss_clip": 0.01156324,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.04475129,
      "balance_loss_mlp": 1.01918864,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 3.3126529453351843,
      "language_loss": 0.81714386,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83897716,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.5912771224975586
    },
    {
      "auxiliary_loss_clip": 0.01155417,
      "auxiliary_loss_mlp": 0.00901155,
      "balance_loss_clip": 1.04590344,
      "balance_loss_mlp": 1.00063431,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 2.089158426958417,
      "language_loss": 0.79218566,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.81275141,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.5957283973693848
    },
    {
      "auxiliary_loss_clip": 0.01163051,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.04738164,
      "balance_loss_mlp": 1.01980507,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 1.8757374199204397,
      "language_loss": 0.73452187,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75643086,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.6200222969055176
    },
    {
      "auxiliary_loss_clip": 0.01149708,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.04740155,
      "balance_loss_mlp": 1.02022946,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.0618654820059357,
      "language_loss": 0.86624396,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88802451,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.681136131286621
    },
    {
      "auxiliary_loss_clip": 0.01144234,
      "auxiliary_loss_mlp": 0.01023397,
      "balance_loss_clip": 1.0447824,
      "balance_loss_mlp": 1.01632488,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.7584528168449378,
      "language_loss": 0.79270697,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81438321,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 3.6523001194000244
    },
    {
      "auxiliary_loss_clip": 0.01152674,
      "auxiliary_loss_mlp": 0.01025258,
      "balance_loss_clip": 1.04764318,
      "balance_loss_mlp": 1.01809382,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.8064762383305744,
      "language_loss": 0.76104057,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78281993,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.627474308013916
    },
    {
      "auxiliary_loss_clip": 0.01161852,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.05208707,
      "balance_loss_mlp": 1.01819015,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 7.045662224936796,
      "language_loss": 0.74402171,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76590681,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.6531364917755127
    },
    {
      "auxiliary_loss_clip": 0.01133203,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.04240453,
      "balance_loss_mlp": 1.02172017,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.3028477845815205,
      "language_loss": 0.90575087,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92737651,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.6866114139556885
    },
    {
      "auxiliary_loss_clip": 0.01130533,
      "auxiliary_loss_mlp": 0.01030916,
      "balance_loss_clip": 1.04634392,
      "balance_loss_mlp": 1.02288175,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 1.9659311806531703,
      "language_loss": 0.77426434,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79587883,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.7245469093322754
    },
    {
      "auxiliary_loss_clip": 0.011423,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.04379225,
      "balance_loss_mlp": 1.01781631,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 2.1021597308599493,
      "language_loss": 0.78518069,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80685472,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.6773478984832764
    },
    {
      "auxiliary_loss_clip": 0.01163466,
      "auxiliary_loss_mlp": 0.01026663,
      "balance_loss_clip": 1.04781759,
      "balance_loss_mlp": 1.01905465,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 5.10972917798328,
      "language_loss": 0.73561043,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75751173,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.6525704860687256
    },
    {
      "auxiliary_loss_clip": 0.01166622,
      "auxiliary_loss_mlp": 0.01025958,
      "balance_loss_clip": 1.04938293,
      "balance_loss_mlp": 1.01836443,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 2.4000217395765304,
      "language_loss": 0.7191416,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74106741,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.830207347869873
    },
    {
      "auxiliary_loss_clip": 0.01151576,
      "auxiliary_loss_mlp": 0.01022476,
      "balance_loss_clip": 1.0457232,
      "balance_loss_mlp": 1.01547503,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 1.8320177103953528,
      "language_loss": 0.81825829,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.83999884,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 3.556673765182495
    },
    {
      "auxiliary_loss_clip": 0.01155516,
      "auxiliary_loss_mlp": 0.01024556,
      "balance_loss_clip": 1.04681969,
      "balance_loss_mlp": 1.01708734,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.9079517088439535,
      "language_loss": 0.69814634,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.71994704,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.5940849781036377
    },
    {
      "auxiliary_loss_clip": 0.01174753,
      "auxiliary_loss_mlp": 0.01026389,
      "balance_loss_clip": 1.05018544,
      "balance_loss_mlp": 1.0183897,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 2.6107061495753,
      "language_loss": 0.81796443,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83997583,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.556520700454712
    },
    {
      "auxiliary_loss_clip": 0.01142589,
      "auxiliary_loss_mlp": 0.01024529,
      "balance_loss_clip": 1.04663897,
      "balance_loss_mlp": 1.01718855,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.0114218411447995,
      "language_loss": 0.87492424,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89659548,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.75632381439209
    },
    {
      "auxiliary_loss_clip": 0.01153783,
      "auxiliary_loss_mlp": 0.01027712,
      "balance_loss_clip": 1.04921937,
      "balance_loss_mlp": 1.01965404,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 4.06646666679954,
      "language_loss": 0.7217986,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74361354,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.6828653812408447
    },
    {
      "auxiliary_loss_clip": 0.01129609,
      "auxiliary_loss_mlp": 0.01021358,
      "balance_loss_clip": 1.04447663,
      "balance_loss_mlp": 1.0144794,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.6777993387136674,
      "language_loss": 0.78296018,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80446982,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.716308116912842
    },
    {
      "auxiliary_loss_clip": 0.01139852,
      "auxiliary_loss_mlp": 0.01022324,
      "balance_loss_clip": 1.046386,
      "balance_loss_mlp": 1.01520097,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.301670905244517,
      "language_loss": 0.84604973,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.86767149,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.6531949043273926
    },
    {
      "auxiliary_loss_clip": 0.01054842,
      "auxiliary_loss_mlp": 0.01001633,
      "balance_loss_clip": 1.01375604,
      "balance_loss_mlp": 1.00084579,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7621130578589598,
      "language_loss": 0.58349401,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60405874,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3077285289764404
    },
    {
      "auxiliary_loss_clip": 0.01149276,
      "auxiliary_loss_mlp": 0.01027251,
      "balance_loss_clip": 1.04477096,
      "balance_loss_mlp": 1.01967514,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 2.3511513909235626,
      "language_loss": 0.65916002,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68092525,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.6579020023345947
    },
    {
      "auxiliary_loss_clip": 0.01162859,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.04705572,
      "balance_loss_mlp": 1.01882851,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.7002153607175883,
      "language_loss": 0.79063565,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.8125242,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 3.6120309829711914
    },
    {
      "auxiliary_loss_clip": 0.01167186,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.04905915,
      "balance_loss_mlp": 1.01929045,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 2.180658889004389,
      "language_loss": 0.78936791,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81130993,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.6287481784820557
    },
    {
      "auxiliary_loss_clip": 0.01164458,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 1.04767025,
      "balance_loss_mlp": 1.01834941,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.2175192577509732,
      "language_loss": 0.81810683,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.8400172,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.614591598510742
    },
    {
      "auxiliary_loss_clip": 0.01141515,
      "auxiliary_loss_mlp": 0.01023735,
      "balance_loss_clip": 1.04091167,
      "balance_loss_mlp": 1.01642501,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 2.4388845857262975,
      "language_loss": 0.74959302,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77124548,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.7311902046203613
    },
    {
      "auxiliary_loss_clip": 0.01160054,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.04711127,
      "balance_loss_mlp": 1.01939964,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.481953635461885,
      "language_loss": 0.85138267,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.87325329,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 3.531571626663208
    },
    {
      "auxiliary_loss_clip": 0.01147908,
      "auxiliary_loss_mlp": 0.0102604,
      "balance_loss_clip": 1.04525304,
      "balance_loss_mlp": 1.01797533,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 2.1293461490133048,
      "language_loss": 0.66936785,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.69110733,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.586301803588867
    },
    {
      "auxiliary_loss_clip": 0.0106539,
      "auxiliary_loss_mlp": 0.01001438,
      "balance_loss_clip": 1.01136684,
      "balance_loss_mlp": 1.00068378,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8250162767471466,
      "language_loss": 0.54484648,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56551474,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.177739381790161
    },
    {
      "auxiliary_loss_clip": 0.01147886,
      "auxiliary_loss_mlp": 0.01023138,
      "balance_loss_clip": 1.04278684,
      "balance_loss_mlp": 1.01581287,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.871848329180586,
      "language_loss": 0.81770134,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.8394115,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.61325740814209
    },
    {
      "auxiliary_loss_clip": 0.01150844,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.04875517,
      "balance_loss_mlp": 1.02034438,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 3.4468147425490026,
      "language_loss": 0.83331144,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85510683,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.7013800144195557
    },
    {
      "auxiliary_loss_clip": 0.01162197,
      "auxiliary_loss_mlp": 0.0102386,
      "balance_loss_clip": 1.04689491,
      "balance_loss_mlp": 1.01670778,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 2.1236237712921047,
      "language_loss": 0.71418267,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73604321,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 3.4670491218566895
    },
    {
      "auxiliary_loss_clip": 0.01174243,
      "auxiliary_loss_mlp": 0.01023075,
      "balance_loss_clip": 1.05212402,
      "balance_loss_mlp": 1.0156244,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 2.0972092511406584,
      "language_loss": 0.74497646,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76694965,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.6867024898529053
    },
    {
      "auxiliary_loss_clip": 0.01171634,
      "auxiliary_loss_mlp": 0.01028114,
      "balance_loss_clip": 1.04998851,
      "balance_loss_mlp": 1.02104521,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.6427518567342,
      "language_loss": 0.72456199,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.7465595,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.4975006580352783
    },
    {
      "auxiliary_loss_clip": 0.01150027,
      "auxiliary_loss_mlp": 0.01031531,
      "balance_loss_clip": 1.04517746,
      "balance_loss_mlp": 1.02338862,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.6725263497289065,
      "language_loss": 0.74539316,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.7672087,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.6577770709991455
    },
    {
      "auxiliary_loss_clip": 0.01156947,
      "auxiliary_loss_mlp": 0.01026969,
      "balance_loss_clip": 1.04524684,
      "balance_loss_mlp": 1.01975036,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 2.212517835226636,
      "language_loss": 0.78725153,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80909061,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.6491127014160156
    },
    {
      "auxiliary_loss_clip": 0.01138479,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.0429728,
      "balance_loss_mlp": 1.02106714,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 3.1460581957158382,
      "language_loss": 0.71860945,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.74028134,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.649221658706665
    },
    {
      "auxiliary_loss_clip": 0.01157996,
      "auxiliary_loss_mlp": 0.01027804,
      "balance_loss_clip": 1.04716992,
      "balance_loss_mlp": 1.01988554,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.317176274365623,
      "language_loss": 0.79525411,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.81711209,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.6493351459503174
    },
    {
      "auxiliary_loss_clip": 0.01161511,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.04902601,
      "balance_loss_mlp": 1.01739872,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 4.029043031511009,
      "language_loss": 0.71160865,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.73347044,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.5815157890319824
    },
    {
      "auxiliary_loss_clip": 0.01147238,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.0448153,
      "balance_loss_mlp": 1.02001333,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 2.2272325342248194,
      "language_loss": 0.78647572,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80822355,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.676560878753662
    },
    {
      "auxiliary_loss_clip": 0.0117005,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 1.04893684,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 2.0656953635970576,
      "language_loss": 0.73285609,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75480324,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 3.4720566272735596
    },
    {
      "auxiliary_loss_clip": 0.01159093,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.04731214,
      "balance_loss_mlp": 1.0173043,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.1587967341970424,
      "language_loss": 0.78726053,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.80909967,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.5508737564086914
    },
    {
      "auxiliary_loss_clip": 0.01160278,
      "auxiliary_loss_mlp": 0.01023585,
      "balance_loss_clip": 1.04894567,
      "balance_loss_mlp": 1.01625347,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.0076692807563035,
      "language_loss": 0.78163034,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.80346894,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.52805233001709
    },
    {
      "auxiliary_loss_clip": 0.01141456,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04053044,
      "balance_loss_mlp": 1.01660895,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 1.9653988351941105,
      "language_loss": 0.72737044,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.74902624,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.795729637145996
    },
    {
      "auxiliary_loss_clip": 0.01160833,
      "auxiliary_loss_mlp": 0.01023833,
      "balance_loss_clip": 1.04771745,
      "balance_loss_mlp": 1.01590252,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.0564320713305335,
      "language_loss": 0.77747381,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.79932046,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.5536081790924072
    },
    {
      "auxiliary_loss_clip": 0.01148931,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.0476824,
      "balance_loss_mlp": 1.02172351,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 2.0714553376867273,
      "language_loss": 0.83603656,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.85781968,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.7108330726623535
    },
    {
      "auxiliary_loss_clip": 0.01167704,
      "auxiliary_loss_mlp": 0.01021986,
      "balance_loss_clip": 1.04692292,
      "balance_loss_mlp": 1.01447606,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 2.319012201665896,
      "language_loss": 0.74134237,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76323926,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.6707382202148438
    },
    {
      "auxiliary_loss_clip": 0.01168526,
      "auxiliary_loss_mlp": 0.01028743,
      "balance_loss_clip": 1.05297995,
      "balance_loss_mlp": 1.02082193,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 2.2130882736706257,
      "language_loss": 0.80224299,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82421565,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.600738763809204
    },
    {
      "auxiliary_loss_clip": 0.01145676,
      "auxiliary_loss_mlp": 0.01025438,
      "balance_loss_clip": 1.047719,
      "balance_loss_mlp": 1.01756406,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.317116166568293,
      "language_loss": 0.77572513,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79743624,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 3.6297690868377686
    },
    {
      "auxiliary_loss_clip": 0.01162282,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.04906011,
      "balance_loss_mlp": 1.02296638,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 2.370906295901337,
      "language_loss": 0.77401125,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79594296,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.631194829940796
    },
    {
      "auxiliary_loss_clip": 0.01164651,
      "auxiliary_loss_mlp": 0.01022415,
      "balance_loss_clip": 1.0512048,
      "balance_loss_mlp": 1.01500607,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 2.3227648820570703,
      "language_loss": 0.74679744,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76866806,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.6768369674682617
    },
    {
      "auxiliary_loss_clip": 0.01161135,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04894114,
      "balance_loss_mlp": 1.01550972,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 2.0776402344611387,
      "language_loss": 0.7531206,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77496213,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.675147533416748
    },
    {
      "auxiliary_loss_clip": 0.01157392,
      "auxiliary_loss_mlp": 0.00901029,
      "balance_loss_clip": 1.04589462,
      "balance_loss_mlp": 1.00070834,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 2.0109715526443015,
      "language_loss": 0.67214036,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.69272459,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 3.604893445968628
    },
    {
      "auxiliary_loss_clip": 0.0117772,
      "auxiliary_loss_mlp": 0.01023273,
      "balance_loss_clip": 1.05409169,
      "balance_loss_mlp": 1.01648688,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.5319311212717615,
      "language_loss": 0.77011472,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79212463,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.58864164352417
    },
    {
      "auxiliary_loss_clip": 0.01159699,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.04883945,
      "balance_loss_mlp": 1.02216411,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 3.082318986612318,
      "language_loss": 0.78477871,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80667001,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.6357052326202393
    },
    {
      "auxiliary_loss_clip": 0.01143368,
      "auxiliary_loss_mlp": 0.01030577,
      "balance_loss_clip": 1.04382527,
      "balance_loss_mlp": 1.02305484,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.1391219302519655,
      "language_loss": 0.75945234,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.78119189,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.6334822177886963
    },
    {
      "auxiliary_loss_clip": 0.01154656,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 1.04700685,
      "balance_loss_mlp": 1.01446009,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.6454518006455674,
      "language_loss": 0.7577644,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77953434,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.6622254848480225
    },
    {
      "auxiliary_loss_clip": 0.0116191,
      "auxiliary_loss_mlp": 0.01024349,
      "balance_loss_clip": 1.05106616,
      "balance_loss_mlp": 1.01658273,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 3.7614175026963506,
      "language_loss": 0.78438509,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80624771,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.617788314819336
    },
    {
      "auxiliary_loss_clip": 0.01146954,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.04864156,
      "balance_loss_mlp": 1.01908505,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 11.479575887506849,
      "language_loss": 0.66917449,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69091034,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 3.6398892402648926
    },
    {
      "auxiliary_loss_clip": 0.0117483,
      "auxiliary_loss_mlp": 0.00901341,
      "balance_loss_clip": 1.04981697,
      "balance_loss_mlp": 1.00078905,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 3.1791200512339497,
      "language_loss": 0.6621424,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68290406,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.5533478260040283
    },
    {
      "auxiliary_loss_clip": 0.01162563,
      "auxiliary_loss_mlp": 0.01021782,
      "balance_loss_clip": 1.04468751,
      "balance_loss_mlp": 1.0143795,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.5040233796587943,
      "language_loss": 0.68168998,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70353341,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.5798799991607666
    },
    {
      "auxiliary_loss_clip": 0.01147571,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 1.04412973,
      "balance_loss_mlp": 1.01505828,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 1.9279039378977445,
      "language_loss": 0.72704661,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.7487514,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.678269386291504
    },
    {
      "auxiliary_loss_clip": 0.01152134,
      "auxiliary_loss_mlp": 0.01027205,
      "balance_loss_clip": 1.04797912,
      "balance_loss_mlp": 1.01990902,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.6346264919967768,
      "language_loss": 0.80018556,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82197899,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.7161977291107178
    },
    {
      "auxiliary_loss_clip": 0.01064002,
      "auxiliary_loss_mlp": 0.01000676,
      "balance_loss_clip": 1.01171947,
      "balance_loss_mlp": 0.99988919,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.700393465686502,
      "language_loss": 0.58170742,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60235423,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.2328567504882812
    },
    {
      "auxiliary_loss_clip": 0.01147414,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.04556692,
      "balance_loss_mlp": 1.02101517,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 1.9841195310244446,
      "language_loss": 0.81260216,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83436358,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.608691453933716
    },
    {
      "auxiliary_loss_clip": 0.01153151,
      "auxiliary_loss_mlp": 0.01026024,
      "balance_loss_clip": 1.04882479,
      "balance_loss_mlp": 1.01905048,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.7880486445751285,
      "language_loss": 0.68621922,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.70801091,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.744250774383545
    },
    {
      "auxiliary_loss_clip": 0.01157916,
      "auxiliary_loss_mlp": 0.01030349,
      "balance_loss_clip": 1.0486877,
      "balance_loss_mlp": 1.02277935,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 3.0241814250596373,
      "language_loss": 0.75384778,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77573049,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.664870023727417
    },
    {
      "auxiliary_loss_clip": 0.01173858,
      "auxiliary_loss_mlp": 0.01031214,
      "balance_loss_clip": 1.05065322,
      "balance_loss_mlp": 1.02342391,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.802924268810993,
      "language_loss": 0.72047073,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74252146,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 3.4224295616149902
    },
    {
      "auxiliary_loss_clip": 0.01153732,
      "auxiliary_loss_mlp": 0.01027347,
      "balance_loss_clip": 1.04740334,
      "balance_loss_mlp": 1.02010489,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 1.9385324138654483,
      "language_loss": 0.8042531,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82606393,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.6762025356292725
    },
    {
      "auxiliary_loss_clip": 0.01166077,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.05125928,
      "balance_loss_mlp": 1.01858497,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 2.885136814946521,
      "language_loss": 0.70109618,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72301793,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.750262498855591
    },
    {
      "auxiliary_loss_clip": 0.01163402,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.05161834,
      "balance_loss_mlp": 1.0192343,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 3.000820014814929,
      "language_loss": 0.74637759,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76827824,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.631450653076172
    },
    {
      "auxiliary_loss_clip": 0.01167537,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.04945874,
      "balance_loss_mlp": 1.01859593,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.4530634301879894,
      "language_loss": 0.71479559,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73673701,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.640962600708008
    },
    {
      "auxiliary_loss_clip": 0.01144468,
      "auxiliary_loss_mlp": 0.01028976,
      "balance_loss_clip": 1.04570377,
      "balance_loss_mlp": 1.02200866,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.7204526720760156,
      "language_loss": 0.7467925,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76852691,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.7121469974517822
    },
    {
      "auxiliary_loss_clip": 0.01133703,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 1.04639149,
      "balance_loss_mlp": 1.01543784,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.039727675645543,
      "language_loss": 0.68603814,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70760202,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.6688973903656006
    },
    {
      "auxiliary_loss_clip": 0.01132382,
      "auxiliary_loss_mlp": 0.01021764,
      "balance_loss_clip": 1.04271734,
      "balance_loss_mlp": 1.01464438,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.4114649586663757,
      "language_loss": 0.6579299,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67947131,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.668903350830078
    },
    {
      "auxiliary_loss_clip": 0.01170135,
      "auxiliary_loss_mlp": 0.01019427,
      "balance_loss_clip": 1.04968929,
      "balance_loss_mlp": 1.01173186,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.1958808473291254,
      "language_loss": 0.76538998,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78728557,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 3.538830518722534
    },
    {
      "auxiliary_loss_clip": 0.01152452,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.04324841,
      "balance_loss_mlp": 1.01740873,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 3.3424044601325087,
      "language_loss": 0.73673654,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75851154,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.567103862762451
    },
    {
      "auxiliary_loss_clip": 0.01140773,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.0423789,
      "balance_loss_mlp": 1.02080035,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.1834524040746204,
      "language_loss": 0.72025681,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.7419582,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.6866679191589355
    },
    {
      "auxiliary_loss_clip": 0.01163551,
      "auxiliary_loss_mlp": 0.01027667,
      "balance_loss_clip": 1.0494616,
      "balance_loss_mlp": 1.02012682,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.8472289629819945,
      "language_loss": 0.69438446,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71629661,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.6097934246063232
    },
    {
      "auxiliary_loss_clip": 0.01158161,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.0480212,
      "balance_loss_mlp": 1.01753068,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 2.068688509779731,
      "language_loss": 0.78277892,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80460906,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 3.551126718521118
    },
    {
      "auxiliary_loss_clip": 0.01155135,
      "auxiliary_loss_mlp": 0.01025627,
      "balance_loss_clip": 1.04698086,
      "balance_loss_mlp": 1.01782441,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.0880324450331784,
      "language_loss": 0.74024504,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76205266,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 2.6254422664642334
    },
    {
      "auxiliary_loss_clip": 0.01148846,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04450762,
      "balance_loss_mlp": 1.01725984,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 1.958701612087198,
      "language_loss": 0.72126591,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74300814,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.7112843990325928
    },
    {
      "auxiliary_loss_clip": 0.01164031,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.04596102,
      "balance_loss_mlp": 1.01744163,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 2.0136498820262347,
      "language_loss": 0.8057009,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.82760054,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.858330488204956
    },
    {
      "auxiliary_loss_clip": 0.01171582,
      "auxiliary_loss_mlp": 0.01021106,
      "balance_loss_clip": 1.05009294,
      "balance_loss_mlp": 1.01426649,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.7769280690284077,
      "language_loss": 0.71998554,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.74191248,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.5709142684936523
    },
    {
      "auxiliary_loss_clip": 0.0113441,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 1.04133725,
      "balance_loss_mlp": 1.020033,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 2.9497114002354974,
      "language_loss": 0.69427443,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71588969,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.751636505126953
    },
    {
      "auxiliary_loss_clip": 0.01147166,
      "auxiliary_loss_mlp": 0.01027768,
      "balance_loss_clip": 1.0415355,
      "balance_loss_mlp": 1.02040648,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 2.6851274239578378,
      "language_loss": 0.84150863,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.863258,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 3.5358691215515137
    },
    {
      "auxiliary_loss_clip": 0.011425,
      "auxiliary_loss_mlp": 0.01024952,
      "balance_loss_clip": 1.04644549,
      "balance_loss_mlp": 1.01731706,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 1.906534298868299,
      "language_loss": 0.66630101,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.68797553,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.6207971572875977
    },
    {
      "auxiliary_loss_clip": 0.01156943,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.04729247,
      "balance_loss_mlp": 1.02076685,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.8340338980499762,
      "language_loss": 0.82980686,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85166001,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.628833532333374
    },
    {
      "auxiliary_loss_clip": 0.01166099,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.04807198,
      "balance_loss_mlp": 1.01743746,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 3.7624936939904403,
      "language_loss": 0.80927503,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.83119059,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.635230541229248
    },
    {
      "auxiliary_loss_clip": 0.01167124,
      "auxiliary_loss_mlp": 0.01027276,
      "balance_loss_clip": 1.04850817,
      "balance_loss_mlp": 1.01933086,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.6673013652226214,
      "language_loss": 0.7689926,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.79093659,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.644585609436035
    },
    {
      "auxiliary_loss_clip": 0.01144446,
      "auxiliary_loss_mlp": 0.01023938,
      "balance_loss_clip": 1.04612923,
      "balance_loss_mlp": 1.01702976,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 2.3376513000092256,
      "language_loss": 0.70157444,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72325832,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.6168625354766846
    },
    {
      "auxiliary_loss_clip": 0.01142351,
      "auxiliary_loss_mlp": 0.0102559,
      "balance_loss_clip": 1.04566979,
      "balance_loss_mlp": 1.01765037,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 2.192789525963291,
      "language_loss": 0.72908688,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.75076628,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.6640024185180664
    },
    {
      "auxiliary_loss_clip": 0.01157949,
      "auxiliary_loss_mlp": 0.01030108,
      "balance_loss_clip": 1.04656661,
      "balance_loss_mlp": 1.02263319,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.7919722783302776,
      "language_loss": 0.75708902,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77896959,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.6564583778381348
    },
    {
      "auxiliary_loss_clip": 0.01163945,
      "auxiliary_loss_mlp": 0.01024531,
      "balance_loss_clip": 1.04701507,
      "balance_loss_mlp": 1.01615036,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.4073364900208647,
      "language_loss": 0.61037171,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63225645,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 3.5775370597839355
    },
    {
      "auxiliary_loss_clip": 0.01160958,
      "auxiliary_loss_mlp": 0.0102136,
      "balance_loss_clip": 1.04504132,
      "balance_loss_mlp": 1.01423466,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.7611808001628058,
      "language_loss": 0.78069299,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80251622,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.634505033493042
    },
    {
      "auxiliary_loss_clip": 0.01140152,
      "auxiliary_loss_mlp": 0.01023321,
      "balance_loss_clip": 1.04167974,
      "balance_loss_mlp": 1.01588821,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 4.367701298026579,
      "language_loss": 0.82531512,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84694993,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.6864705085754395
    },
    {
      "auxiliary_loss_clip": 0.01164426,
      "auxiliary_loss_mlp": 0.0102964,
      "balance_loss_clip": 1.0468626,
      "balance_loss_mlp": 1.02268445,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.8717318531010252,
      "language_loss": 0.72017908,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74211973,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.6735000610351562
    },
    {
      "auxiliary_loss_clip": 0.01128471,
      "auxiliary_loss_mlp": 0.01038006,
      "balance_loss_clip": 1.04185784,
      "balance_loss_mlp": 1.03037655,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 2.4934979782775497,
      "language_loss": 0.63722223,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.65888703,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.867217779159546
    },
    {
      "auxiliary_loss_clip": 0.0106339,
      "auxiliary_loss_mlp": 0.01001835,
      "balance_loss_clip": 1.01407802,
      "balance_loss_mlp": 1.00103593,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7465246087656756,
      "language_loss": 0.58707178,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60772401,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.103949785232544
    },
    {
      "auxiliary_loss_clip": 0.01178429,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.05231166,
      "balance_loss_mlp": 1.0186491,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.7532686787823106,
      "language_loss": 0.88832891,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91037667,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.6073286533355713
    },
    {
      "auxiliary_loss_clip": 0.01163101,
      "auxiliary_loss_mlp": 0.01030163,
      "balance_loss_clip": 1.04936171,
      "balance_loss_mlp": 1.021896,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.7931755011820067,
      "language_loss": 0.68541157,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70734423,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.6531243324279785
    },
    {
      "auxiliary_loss_clip": 0.01172983,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 1.04989052,
      "balance_loss_mlp": 1.01830101,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.3576044122948847,
      "language_loss": 0.76705259,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78903896,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.555051803588867
    },
    {
      "auxiliary_loss_clip": 0.01158448,
      "auxiliary_loss_mlp": 0.01024013,
      "balance_loss_clip": 1.04758739,
      "balance_loss_mlp": 1.01653838,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 3.2365108180833713,
      "language_loss": 0.72581005,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74763459,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 3.645190715789795
    },
    {
      "auxiliary_loss_clip": 0.01161286,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.04972911,
      "balance_loss_mlp": 1.02239609,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 2.511716532710839,
      "language_loss": 0.77811468,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.8000232,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.759556770324707
    },
    {
      "auxiliary_loss_clip": 0.01171318,
      "auxiliary_loss_mlp": 0.00901539,
      "balance_loss_clip": 1.04902887,
      "balance_loss_mlp": 1.00078535,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.1712485745043093,
      "language_loss": 0.71778905,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73851758,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.52105712890625
    },
    {
      "auxiliary_loss_clip": 0.01069133,
      "auxiliary_loss_mlp": 0.010017,
      "balance_loss_clip": 1.01363802,
      "balance_loss_mlp": 1.0009129,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7659381410148817,
      "language_loss": 0.64231902,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66302735,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 4.270172357559204
    },
    {
      "auxiliary_loss_clip": 0.01140006,
      "auxiliary_loss_mlp": 0.01027525,
      "balance_loss_clip": 1.04473615,
      "balance_loss_mlp": 1.02001536,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 2.1145603547043192,
      "language_loss": 0.70909059,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.73076588,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.6992263793945312
    },
    {
      "auxiliary_loss_clip": 0.01157927,
      "auxiliary_loss_mlp": 0.01030015,
      "balance_loss_clip": 1.04458857,
      "balance_loss_mlp": 1.02255821,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 2.152326097656261,
      "language_loss": 0.70654035,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72841978,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.583289861679077
    },
    {
      "auxiliary_loss_clip": 0.01149163,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.04724419,
      "balance_loss_mlp": 1.01779473,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 2.542108764327993,
      "language_loss": 0.75007331,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77181518,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.7346203327178955
    },
    {
      "auxiliary_loss_clip": 0.01164215,
      "auxiliary_loss_mlp": 0.01033408,
      "balance_loss_clip": 1.04782963,
      "balance_loss_mlp": 1.02487254,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.3537039834499014,
      "language_loss": 0.71335661,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73533285,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.658755302429199
    },
    {
      "auxiliary_loss_clip": 0.01162362,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.04919958,
      "balance_loss_mlp": 1.01719737,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.7455104423938772,
      "language_loss": 0.79893279,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.82080299,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.6832456588745117
    },
    {
      "auxiliary_loss_clip": 0.01151327,
      "auxiliary_loss_mlp": 0.01024555,
      "balance_loss_clip": 1.04736364,
      "balance_loss_mlp": 1.01723015,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 3.4124712704588434,
      "language_loss": 0.80349708,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82525587,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 3.503981351852417
    },
    {
      "auxiliary_loss_clip": 0.01054645,
      "auxiliary_loss_mlp": 0.01000749,
      "balance_loss_clip": 1.01007032,
      "balance_loss_mlp": 0.99995035,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.8250538294649853,
      "language_loss": 0.57987976,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60043371,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.1969878673553467
    },
    {
      "auxiliary_loss_clip": 0.01163003,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.04698133,
      "balance_loss_mlp": 1.01746726,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 2.17017005095105,
      "language_loss": 0.62463152,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64651972,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.682387590408325
    },
    {
      "auxiliary_loss_clip": 0.01058018,
      "auxiliary_loss_mlp": 0.00891285,
      "balance_loss_clip": 1.01331091,
      "balance_loss_mlp": 1.00000119,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.8455883933239061,
      "language_loss": 0.52409178,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54358482,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.195159435272217
    },
    {
      "auxiliary_loss_clip": 0.01163517,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.05019891,
      "balance_loss_mlp": 1.02134395,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 3.316790208348038,
      "language_loss": 0.71890163,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.7408269,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.6512537002563477
    },
    {
      "auxiliary_loss_clip": 0.01066214,
      "auxiliary_loss_mlp": 0.01000446,
      "balance_loss_clip": 1.01261449,
      "balance_loss_mlp": 0.99960595,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.665995161714272,
      "language_loss": 0.60966635,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63033295,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.179468870162964
    },
    {
      "auxiliary_loss_clip": 0.01163554,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.05029047,
      "balance_loss_mlp": 1.02226186,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.0978860488366333,
      "language_loss": 0.70036155,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.72230029,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.6418633460998535
    },
    {
      "auxiliary_loss_clip": 0.01172802,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.04840636,
      "balance_loss_mlp": 1.01787996,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.3655271466600563,
      "language_loss": 0.77884638,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.80082899,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.547790050506592
    },
    {
      "auxiliary_loss_clip": 0.01164536,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 1.05271316,
      "balance_loss_mlp": 1.02110529,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.7640811003017574,
      "language_loss": 0.71739173,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.73932153,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 3.52144193649292
    },
    {
      "auxiliary_loss_clip": 0.01167165,
      "auxiliary_loss_mlp": 0.01027117,
      "balance_loss_clip": 1.04901648,
      "balance_loss_mlp": 1.01962459,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 2.5581138212849943,
      "language_loss": 0.82759559,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84953839,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.565000534057617
    },
    {
      "auxiliary_loss_clip": 0.01150669,
      "auxiliary_loss_mlp": 0.01026006,
      "balance_loss_clip": 1.04722929,
      "balance_loss_mlp": 1.01889491,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.7291465992983472,
      "language_loss": 0.78082806,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80259478,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.6560401916503906
    },
    {
      "auxiliary_loss_clip": 0.01147559,
      "auxiliary_loss_mlp": 0.01026583,
      "balance_loss_clip": 1.04685307,
      "balance_loss_mlp": 1.01876235,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 2.1279724222662524,
      "language_loss": 0.67518443,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69692576,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.773576259613037
    },
    {
      "auxiliary_loss_clip": 0.0106364,
      "auxiliary_loss_mlp": 0.01001352,
      "balance_loss_clip": 1.0140245,
      "balance_loss_mlp": 1.00059831,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8090622077184527,
      "language_loss": 0.67572224,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69637215,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.2540833950042725
    },
    {
      "auxiliary_loss_clip": 0.01176246,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 1.05254769,
      "balance_loss_mlp": 1.01858306,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 2.020588549814821,
      "language_loss": 0.74004757,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76207471,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.6405646800994873
    },
    {
      "auxiliary_loss_clip": 0.01167629,
      "auxiliary_loss_mlp": 0.01027416,
      "balance_loss_clip": 1.04897583,
      "balance_loss_mlp": 1.0190239,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.696290617491429,
      "language_loss": 0.71271813,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73466861,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.6564884185791016
    },
    {
      "auxiliary_loss_clip": 0.01062925,
      "auxiliary_loss_mlp": 0.01002034,
      "balance_loss_clip": 1.01353407,
      "balance_loss_mlp": 1.00125313,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.673835425557958,
      "language_loss": 0.59745467,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61810422,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.233822822570801
    },
    {
      "auxiliary_loss_clip": 0.0114146,
      "auxiliary_loss_mlp": 0.01026063,
      "balance_loss_clip": 1.04522157,
      "balance_loss_mlp": 1.018085,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 4.314849946411502,
      "language_loss": 0.62467974,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.64635497,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.7712080478668213
    },
    {
      "auxiliary_loss_clip": 0.01137493,
      "auxiliary_loss_mlp": 0.0102501,
      "balance_loss_clip": 1.0411582,
      "balance_loss_mlp": 1.01752353,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 2.104707218817589,
      "language_loss": 0.82173228,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84335726,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.6390910148620605
    },
    {
      "auxiliary_loss_clip": 0.01179839,
      "auxiliary_loss_mlp": 0.01023759,
      "balance_loss_clip": 1.05147362,
      "balance_loss_mlp": 1.01523578,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.9651325836940756,
      "language_loss": 0.85107642,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87311238,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.6012330055236816
    },
    {
      "auxiliary_loss_clip": 0.01164122,
      "auxiliary_loss_mlp": 0.01023371,
      "balance_loss_clip": 1.04845798,
      "balance_loss_mlp": 1.01594138,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 2.3180773507649852,
      "language_loss": 0.74883121,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77070618,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.5911099910736084
    },
    {
      "auxiliary_loss_clip": 0.01147644,
      "auxiliary_loss_mlp": 0.01028051,
      "balance_loss_clip": 1.04296112,
      "balance_loss_mlp": 1.02012372,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 1.7983899299039325,
      "language_loss": 0.69444329,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71620023,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 3.7500698566436768
    },
    {
      "auxiliary_loss_clip": 0.01144425,
      "auxiliary_loss_mlp": 0.01023101,
      "balance_loss_clip": 1.0430429,
      "balance_loss_mlp": 1.0157876,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.920272946326938,
      "language_loss": 0.79689145,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81856668,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.7109570503234863
    },
    {
      "auxiliary_loss_clip": 0.01128862,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.04059315,
      "balance_loss_mlp": 1.02124035,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.8553405056194858,
      "language_loss": 0.74737322,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76895446,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.766935110092163
    },
    {
      "auxiliary_loss_clip": 0.01162092,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.04628325,
      "balance_loss_mlp": 1.02057314,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 2.3626229715149876,
      "language_loss": 0.7254014,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74729908,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.5532429218292236
    },
    {
      "auxiliary_loss_clip": 0.01154228,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.04584599,
      "balance_loss_mlp": 1.01956594,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.925449258774216,
      "language_loss": 0.81132025,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.83312988,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.480534791946411
    },
    {
      "auxiliary_loss_clip": 0.01165814,
      "auxiliary_loss_mlp": 0.01023918,
      "balance_loss_clip": 1.04912269,
      "balance_loss_mlp": 1.01690245,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 1.7174277800997948,
      "language_loss": 0.78033316,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80223048,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.485267400741577
    },
    {
      "auxiliary_loss_clip": 0.01166515,
      "auxiliary_loss_mlp": 0.01024833,
      "balance_loss_clip": 1.04816866,
      "balance_loss_mlp": 1.01669133,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 2.1732360723893267,
      "language_loss": 0.74027765,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76219118,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.545966625213623
    },
    {
      "auxiliary_loss_clip": 0.01168067,
      "auxiliary_loss_mlp": 0.01028123,
      "balance_loss_clip": 1.05029643,
      "balance_loss_mlp": 1.02010012,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 2.133475442948772,
      "language_loss": 0.83061111,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85257298,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 3.3258702754974365
    },
    {
      "auxiliary_loss_clip": 0.01071895,
      "auxiliary_loss_mlp": 0.01001104,
      "balance_loss_clip": 1.01204789,
      "balance_loss_mlp": 1.00030553,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8451292923351207,
      "language_loss": 0.66280901,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68353891,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.0162034034729004
    },
    {
      "auxiliary_loss_clip": 0.01174505,
      "auxiliary_loss_mlp": 0.01024299,
      "balance_loss_clip": 1.05096555,
      "balance_loss_mlp": 1.01672935,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 1.8669561597035207,
      "language_loss": 0.76727486,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78926289,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.6374361515045166
    },
    {
      "auxiliary_loss_clip": 0.01146631,
      "auxiliary_loss_mlp": 0.01025566,
      "balance_loss_clip": 1.04534245,
      "balance_loss_mlp": 1.01815748,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.9072405568755149,
      "language_loss": 0.70615351,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72787547,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.7841529846191406
    },
    {
      "auxiliary_loss_clip": 0.01147621,
      "auxiliary_loss_mlp": 0.01029036,
      "balance_loss_clip": 1.04623854,
      "balance_loss_mlp": 1.02094793,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 2.6436614679652544,
      "language_loss": 0.65180933,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.673576,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.7941174507141113
    },
    {
      "auxiliary_loss_clip": 0.01158252,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.04969859,
      "balance_loss_mlp": 1.01695311,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 1.9459551072786054,
      "language_loss": 0.65406883,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.6758939,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.669551372528076
    },
    {
      "auxiliary_loss_clip": 0.01145055,
      "auxiliary_loss_mlp": 0.01023867,
      "balance_loss_clip": 1.04826546,
      "balance_loss_mlp": 1.01665473,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 1.9779339547411037,
      "language_loss": 0.65021282,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67190206,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.686066150665283
    },
    {
      "auxiliary_loss_clip": 0.01046171,
      "auxiliary_loss_mlp": 0.01001442,
      "balance_loss_clip": 1.01029015,
      "balance_loss_mlp": 1.00058389,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6913778962147126,
      "language_loss": 0.51328766,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53376383,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.2492878437042236
    },
    {
      "auxiliary_loss_clip": 0.01152089,
      "auxiliary_loss_mlp": 0.0102301,
      "balance_loss_clip": 1.0470531,
      "balance_loss_mlp": 1.01518381,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 6.879334178651114,
      "language_loss": 0.70989895,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73164999,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 3.533125877380371
    },
    {
      "auxiliary_loss_clip": 0.01160284,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04659569,
      "balance_loss_mlp": 1.01723313,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 1.9903803191850356,
      "language_loss": 0.83529353,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85714483,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.5980982780456543
    },
    {
      "auxiliary_loss_clip": 0.01177697,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.05052876,
      "balance_loss_mlp": 1.01746488,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.6702170805227765,
      "language_loss": 0.84627998,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86830932,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.4727256298065186
    },
    {
      "auxiliary_loss_clip": 0.0116532,
      "auxiliary_loss_mlp": 0.01022453,
      "balance_loss_clip": 1.04831505,
      "balance_loss_mlp": 1.01580405,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.8238416017274226,
      "language_loss": 0.72226632,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74414408,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.5473294258117676
    },
    {
      "auxiliary_loss_clip": 0.01157936,
      "auxiliary_loss_mlp": 0.0102638,
      "balance_loss_clip": 1.0462954,
      "balance_loss_mlp": 1.01906645,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.7856882329704227,
      "language_loss": 0.65111089,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67295408,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.654498815536499
    },
    {
      "auxiliary_loss_clip": 0.01170333,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.05123043,
      "balance_loss_mlp": 1.01778555,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.293522413904327,
      "language_loss": 0.71351087,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73546588,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.5583229064941406
    },
    {
      "auxiliary_loss_clip": 0.01169311,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.04718339,
      "balance_loss_mlp": 1.019696,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.8847086304037044,
      "language_loss": 0.79022026,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81218541,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.501467227935791
    },
    {
      "auxiliary_loss_clip": 0.01166907,
      "auxiliary_loss_mlp": 0.01028875,
      "balance_loss_clip": 1.05003643,
      "balance_loss_mlp": 1.02008891,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 1.8941376886207395,
      "language_loss": 0.73230863,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.7542665,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.626264810562134
    },
    {
      "auxiliary_loss_clip": 0.0116447,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.05201292,
      "balance_loss_mlp": 1.01782632,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 2.2614930218674996,
      "language_loss": 0.72926396,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75116336,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.5510919094085693
    },
    {
      "auxiliary_loss_clip": 0.01162319,
      "auxiliary_loss_mlp": 0.01026619,
      "balance_loss_clip": 1.04845071,
      "balance_loss_mlp": 1.01795268,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.5076505511760088,
      "language_loss": 0.74850267,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.770392,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 3.643019199371338
    },
    {
      "auxiliary_loss_clip": 0.01145878,
      "auxiliary_loss_mlp": 0.01027905,
      "balance_loss_clip": 1.04753423,
      "balance_loss_mlp": 1.01990318,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 2.883958036595955,
      "language_loss": 0.77542233,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79716015,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.643672227859497
    },
    {
      "auxiliary_loss_clip": 0.0115472,
      "auxiliary_loss_mlp": 0.01026944,
      "balance_loss_clip": 1.04665756,
      "balance_loss_mlp": 1.01944256,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.707487817357887,
      "language_loss": 0.79492944,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81674612,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.5843913555145264
    },
    {
      "auxiliary_loss_clip": 0.01165357,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.04975104,
      "balance_loss_mlp": 1.01843309,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 2.3289981570640492,
      "language_loss": 0.69510996,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71702528,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 3.524624824523926
    },
    {
      "auxiliary_loss_clip": 0.0114521,
      "auxiliary_loss_mlp": 0.01022619,
      "balance_loss_clip": 1.04443884,
      "balance_loss_mlp": 1.01471257,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.0102986553901876,
      "language_loss": 0.67015177,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.6918301,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.6514899730682373
    },
    {
      "auxiliary_loss_clip": 0.01171181,
      "auxiliary_loss_mlp": 0.01026716,
      "balance_loss_clip": 1.04741311,
      "balance_loss_mlp": 1.01909804,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.2751579572449043,
      "language_loss": 0.86391973,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88589871,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.7615621089935303
    },
    {
      "auxiliary_loss_clip": 0.01155322,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.04708552,
      "balance_loss_mlp": 1.01943779,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.7175153195788115,
      "language_loss": 0.81237113,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83419424,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.7078568935394287
    },
    {
      "auxiliary_loss_clip": 0.01169835,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.04869664,
      "balance_loss_mlp": 1.02525043,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.8520531032399086,
      "language_loss": 0.7060194,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72804511,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.705975294113159
    },
    {
      "auxiliary_loss_clip": 0.01152053,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.0484755,
      "balance_loss_mlp": 1.02107632,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 2.560534951418322,
      "language_loss": 0.84839034,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.8702035,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.6383309364318848
    },
    {
      "auxiliary_loss_clip": 0.01135621,
      "auxiliary_loss_mlp": 0.01025525,
      "balance_loss_clip": 1.0443542,
      "balance_loss_mlp": 1.01782393,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.029350834345542,
      "language_loss": 0.88343364,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90504509,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 3.609707832336426
    },
    {
      "auxiliary_loss_clip": 0.01174013,
      "auxiliary_loss_mlp": 0.01027704,
      "balance_loss_clip": 1.05122066,
      "balance_loss_mlp": 1.02041125,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.254680395357507,
      "language_loss": 0.78031623,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.80233335,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.482785940170288
    },
    {
      "auxiliary_loss_clip": 0.01155019,
      "auxiliary_loss_mlp": 0.01028881,
      "balance_loss_clip": 1.04710662,
      "balance_loss_mlp": 1.02128994,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.3996243041626886,
      "language_loss": 0.88400161,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90584064,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.5569334030151367
    },
    {
      "auxiliary_loss_clip": 0.01057518,
      "auxiliary_loss_mlp": 0.01000976,
      "balance_loss_clip": 1.0128541,
      "balance_loss_mlp": 1.00026369,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7881049128887455,
      "language_loss": 0.59685481,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61743975,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.256700038909912
    },
    {
      "auxiliary_loss_clip": 0.01163469,
      "auxiliary_loss_mlp": 0.00901257,
      "balance_loss_clip": 1.04979742,
      "balance_loss_mlp": 1.00074077,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.6617493910291719,
      "language_loss": 0.79236281,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.8130101,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.6116416454315186
    },
    {
      "auxiliary_loss_clip": 0.01150472,
      "auxiliary_loss_mlp": 0.01028622,
      "balance_loss_clip": 1.04677606,
      "balance_loss_mlp": 1.02082562,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 1.8894724281740365,
      "language_loss": 0.73145843,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75324941,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.7231438159942627
    },
    {
      "auxiliary_loss_clip": 0.01071509,
      "auxiliary_loss_mlp": 0.01001155,
      "balance_loss_clip": 1.01150799,
      "balance_loss_mlp": 1.00041032,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7135217556168232,
      "language_loss": 0.5897181,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.61044478,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.2367303371429443
    },
    {
      "auxiliary_loss_clip": 0.01148896,
      "auxiliary_loss_mlp": 0.01028105,
      "balance_loss_clip": 1.04475546,
      "balance_loss_mlp": 1.01990318,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 1.9495037616038011,
      "language_loss": 0.6831733,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.7049433,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.6093389987945557
    },
    {
      "auxiliary_loss_clip": 0.01143431,
      "auxiliary_loss_mlp": 0.01022818,
      "balance_loss_clip": 1.04352152,
      "balance_loss_mlp": 1.01497984,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 1.8428609729881222,
      "language_loss": 0.74994707,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77160954,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 3.5377864837646484
    },
    {
      "auxiliary_loss_clip": 0.011238,
      "auxiliary_loss_mlp": 0.01022551,
      "balance_loss_clip": 1.0417769,
      "balance_loss_mlp": 1.01518393,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 2.1116016292099054,
      "language_loss": 0.76243138,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.7838949,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.774101734161377
    },
    {
      "auxiliary_loss_clip": 0.01149541,
      "auxiliary_loss_mlp": 0.01028694,
      "balance_loss_clip": 1.04348326,
      "balance_loss_mlp": 1.01996803,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.3539045688529585,
      "language_loss": 0.69248974,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71427214,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.686192750930786
    },
    {
      "auxiliary_loss_clip": 0.01164471,
      "auxiliary_loss_mlp": 0.01028469,
      "balance_loss_clip": 1.04534566,
      "balance_loss_mlp": 1.02092636,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 2.1920157876753277,
      "language_loss": 0.72627032,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.7481997,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.6224398612976074
    },
    {
      "auxiliary_loss_clip": 0.01139691,
      "auxiliary_loss_mlp": 0.01028372,
      "balance_loss_clip": 1.04332018,
      "balance_loss_mlp": 1.01984847,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 1.8174371732297518,
      "language_loss": 0.80639958,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82808018,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.648690700531006
    },
    {
      "auxiliary_loss_clip": 0.01134868,
      "auxiliary_loss_mlp": 0.01024806,
      "balance_loss_clip": 1.04574704,
      "balance_loss_mlp": 1.01775134,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 2.2795922716761963,
      "language_loss": 0.66601682,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68761349,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.8062515258789062
    },
    {
      "auxiliary_loss_clip": 0.01171071,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.05161572,
      "balance_loss_mlp": 1.02132285,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 1.9691428279961876,
      "language_loss": 0.74310386,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.76510155,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.5944793224334717
    },
    {
      "auxiliary_loss_clip": 0.01157748,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.04677153,
      "balance_loss_mlp": 1.02064514,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.124929080409569,
      "language_loss": 0.80951691,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.83138168,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.6288647651672363
    },
    {
      "auxiliary_loss_clip": 0.01148403,
      "auxiliary_loss_mlp": 0.01028678,
      "balance_loss_clip": 1.05117846,
      "balance_loss_mlp": 1.02131414,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.7105839321924514,
      "language_loss": 0.67174464,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69351542,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.6103570461273193
    },
    {
      "auxiliary_loss_clip": 0.01165915,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.04991078,
      "balance_loss_mlp": 1.01747382,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 2.134114826524512,
      "language_loss": 0.76838958,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.790295,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.570808172225952
    },
    {
      "auxiliary_loss_clip": 0.01151522,
      "auxiliary_loss_mlp": 0.00900992,
      "balance_loss_clip": 1.04692864,
      "balance_loss_mlp": 1.00079882,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 2.2103706238159666,
      "language_loss": 0.711532,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.73205709,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 3.603912591934204
    },
    {
      "auxiliary_loss_clip": 0.01156321,
      "auxiliary_loss_mlp": 0.01022557,
      "balance_loss_clip": 1.04829502,
      "balance_loss_mlp": 1.01492119,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.8552360730946822,
      "language_loss": 0.73629105,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75807983,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.702711582183838
    },
    {
      "auxiliary_loss_clip": 0.01161142,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.04887605,
      "balance_loss_mlp": 1.02141988,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 1.9840812382639699,
      "language_loss": 0.95730466,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97920585,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.6745269298553467
    },
    {
      "auxiliary_loss_clip": 0.01156913,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.04650998,
      "balance_loss_mlp": 1.01824856,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.81499310702997,
      "language_loss": 0.71793783,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.7397629,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 3.5935556888580322
    },
    {
      "auxiliary_loss_clip": 0.0116243,
      "auxiliary_loss_mlp": 0.00901582,
      "balance_loss_clip": 1.04490137,
      "balance_loss_mlp": 1.00086963,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.913060933042711,
      "language_loss": 0.76024961,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78088975,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.8040926456451416
    },
    {
      "auxiliary_loss_clip": 0.011535,
      "auxiliary_loss_mlp": 0.01022585,
      "balance_loss_clip": 1.04340386,
      "balance_loss_mlp": 1.01541424,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.8139148066507436,
      "language_loss": 0.76958996,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79135084,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.7724266052246094
    },
    {
      "auxiliary_loss_clip": 0.01153843,
      "auxiliary_loss_mlp": 0.00901348,
      "balance_loss_clip": 1.04499364,
      "balance_loss_mlp": 1.00083041,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.7657847951364567,
      "language_loss": 0.81039602,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.83094794,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.629667282104492
    },
    {
      "auxiliary_loss_clip": 0.01124776,
      "auxiliary_loss_mlp": 0.00900919,
      "balance_loss_clip": 1.04598475,
      "balance_loss_mlp": 1.00086999,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 2.11680276769467,
      "language_loss": 0.73380125,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75405818,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.7477152347564697
    },
    {
      "auxiliary_loss_clip": 0.01167581,
      "auxiliary_loss_mlp": 0.01022794,
      "balance_loss_clip": 1.04706466,
      "balance_loss_mlp": 1.01564741,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.9940410589173905,
      "language_loss": 0.71414477,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.73604846,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 3.5780582427978516
    },
    {
      "auxiliary_loss_clip": 0.01155273,
      "auxiliary_loss_mlp": 0.00901078,
      "balance_loss_clip": 1.04599106,
      "balance_loss_mlp": 1.00075531,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 2.704231906618897,
      "language_loss": 0.80556256,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.8261261,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.581972122192383
    },
    {
      "auxiliary_loss_clip": 0.01163454,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.05481958,
      "balance_loss_mlp": 1.02355456,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 2.389136511008275,
      "language_loss": 0.76342738,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78537166,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.563234567642212
    },
    {
      "auxiliary_loss_clip": 0.01153284,
      "auxiliary_loss_mlp": 0.01024753,
      "balance_loss_clip": 1.04857755,
      "balance_loss_mlp": 1.01811945,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.7818001254650173,
      "language_loss": 0.7277264,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.74950683,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.6624903678894043
    },
    {
      "auxiliary_loss_clip": 0.0116541,
      "auxiliary_loss_mlp": 0.01021668,
      "balance_loss_clip": 1.04915917,
      "balance_loss_mlp": 1.01432443,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 2.1961534804247864,
      "language_loss": 0.78908741,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.81095815,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.6211843490600586
    },
    {
      "auxiliary_loss_clip": 0.01172977,
      "auxiliary_loss_mlp": 0.01025589,
      "balance_loss_clip": 1.05292606,
      "balance_loss_mlp": 1.01798332,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 1.8383155459622575,
      "language_loss": 0.73346448,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75545013,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.65862774848938
    },
    {
      "auxiliary_loss_clip": 0.01158701,
      "auxiliary_loss_mlp": 0.01024636,
      "balance_loss_clip": 1.04841232,
      "balance_loss_mlp": 1.01653552,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 9.574768207667175,
      "language_loss": 0.75278109,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77461445,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.673610210418701
    },
    {
      "auxiliary_loss_clip": 0.01148512,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04422402,
      "balance_loss_mlp": 1.01642132,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.5905160204104476,
      "language_loss": 0.76977217,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79149985,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.7230582237243652
    },
    {
      "auxiliary_loss_clip": 0.01175085,
      "auxiliary_loss_mlp": 0.01029931,
      "balance_loss_clip": 1.05230832,
      "balance_loss_mlp": 1.02262962,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 2.0141598185117724,
      "language_loss": 0.76851881,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.79056901,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.509829044342041
    },
    {
      "auxiliary_loss_clip": 0.01172085,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.04937983,
      "balance_loss_mlp": 1.01681232,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 2.9532641934109756,
      "language_loss": 0.70605099,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72802246,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 3.5654125213623047
    },
    {
      "auxiliary_loss_clip": 0.01154631,
      "auxiliary_loss_mlp": 0.01024148,
      "balance_loss_clip": 1.04423451,
      "balance_loss_mlp": 1.01712656,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.679510960135097,
      "language_loss": 0.73567879,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75746661,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.693718194961548
    },
    {
      "auxiliary_loss_clip": 0.01168562,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 1.04901791,
      "balance_loss_mlp": 1.01740861,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 2.3835213873525944,
      "language_loss": 0.68577206,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.70771253,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.526322603225708
    },
    {
      "auxiliary_loss_clip": 0.01163827,
      "auxiliary_loss_mlp": 0.01024145,
      "balance_loss_clip": 1.04954851,
      "balance_loss_mlp": 1.01676631,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 2.0064498161226756,
      "language_loss": 0.70952511,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73140484,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.707756519317627
    },
    {
      "auxiliary_loss_clip": 0.01071429,
      "auxiliary_loss_mlp": 0.00890792,
      "balance_loss_clip": 1.01118898,
      "balance_loss_mlp": 0.99997056,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7430088205687284,
      "language_loss": 0.55474871,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57437098,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.193079710006714
    },
    {
      "auxiliary_loss_clip": 0.01145905,
      "auxiliary_loss_mlp": 0.01027317,
      "balance_loss_clip": 1.0477376,
      "balance_loss_mlp": 1.01883483,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.9865026751674852,
      "language_loss": 0.68220878,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70394099,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.7165114879608154
    },
    {
      "auxiliary_loss_clip": 0.01164881,
      "auxiliary_loss_mlp": 0.01020941,
      "balance_loss_clip": 1.04978657,
      "balance_loss_mlp": 1.01379395,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.903737970100496,
      "language_loss": 0.72792912,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.74978727,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.6358373165130615
    },
    {
      "auxiliary_loss_clip": 0.01164736,
      "auxiliary_loss_mlp": 0.01025339,
      "balance_loss_clip": 1.04704738,
      "balance_loss_mlp": 1.01800418,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.7127583627923657,
      "language_loss": 0.70495397,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.7268548,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.6522817611694336
    },
    {
      "auxiliary_loss_clip": 0.0114484,
      "auxiliary_loss_mlp": 0.01022884,
      "balance_loss_clip": 1.04408932,
      "balance_loss_mlp": 1.01510596,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.162680621210382,
      "language_loss": 0.74093038,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76260757,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.6567530632019043
    },
    {
      "auxiliary_loss_clip": 0.01071061,
      "auxiliary_loss_mlp": 0.01002271,
      "balance_loss_clip": 1.01100922,
      "balance_loss_mlp": 1.00149632,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.7036670897671492,
      "language_loss": 0.57912195,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.59985518,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 4.186651945114136
    },
    {
      "auxiliary_loss_clip": 0.01157843,
      "auxiliary_loss_mlp": 0.01026898,
      "balance_loss_clip": 1.04870415,
      "balance_loss_mlp": 1.01974285,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.8421386859141797,
      "language_loss": 0.78895766,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81080508,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.6796700954437256
    },
    {
      "auxiliary_loss_clip": 0.01166375,
      "auxiliary_loss_mlp": 0.01028955,
      "balance_loss_clip": 1.05221868,
      "balance_loss_mlp": 1.02087855,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 3.189698664684235,
      "language_loss": 0.73758948,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.75954276,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.653834581375122
    },
    {
      "auxiliary_loss_clip": 0.01111909,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.03891814,
      "balance_loss_mlp": 1.01904321,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 2.180870202336781,
      "language_loss": 0.75926626,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78064698,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.831723213195801
    },
    {
      "auxiliary_loss_clip": 0.0115458,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.04319704,
      "balance_loss_mlp": 1.01697946,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 1.850236398409986,
      "language_loss": 0.73363733,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75543261,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 3.62827467918396
    },
    {
      "auxiliary_loss_clip": 0.01174391,
      "auxiliary_loss_mlp": 0.01024833,
      "balance_loss_clip": 1.04883075,
      "balance_loss_mlp": 1.01756752,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 2.397049367847009,
      "language_loss": 0.68277931,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70477164,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.6032509803771973
    },
    {
      "auxiliary_loss_clip": 0.01147132,
      "auxiliary_loss_mlp": 0.0103241,
      "balance_loss_clip": 1.04542804,
      "balance_loss_mlp": 1.02536464,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.7581227908198,
      "language_loss": 0.78466052,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80645597,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.799039363861084
    },
    {
      "auxiliary_loss_clip": 0.01148117,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.04170001,
      "balance_loss_mlp": 1.01999211,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 2.0482690232395346,
      "language_loss": 0.64327502,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66503042,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.6760964393615723
    },
    {
      "auxiliary_loss_clip": 0.01151143,
      "auxiliary_loss_mlp": 0.01028878,
      "balance_loss_clip": 1.04593194,
      "balance_loss_mlp": 1.02067626,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.7762253863761126,
      "language_loss": 0.81277019,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83457041,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.687894105911255
    },
    {
      "auxiliary_loss_clip": 0.01144346,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.04497838,
      "balance_loss_mlp": 1.01493597,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 2.244043374104957,
      "language_loss": 0.80012572,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82179141,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 3.619121551513672
    },
    {
      "auxiliary_loss_clip": 0.01164802,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 1.04777765,
      "balance_loss_mlp": 1.02130365,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.0207345735591122,
      "language_loss": 0.75382388,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77576947,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.639937400817871
    },
    {
      "auxiliary_loss_clip": 0.01149053,
      "auxiliary_loss_mlp": 0.0102797,
      "balance_loss_clip": 1.04553556,
      "balance_loss_mlp": 1.02008092,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 1.870418572297052,
      "language_loss": 0.74583858,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76760876,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.704463243484497
    },
    {
      "auxiliary_loss_clip": 0.01164394,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.04837096,
      "balance_loss_mlp": 1.01980376,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 1.9195524521620984,
      "language_loss": 0.86642611,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88834107,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.5361270904541016
    },
    {
      "auxiliary_loss_clip": 0.01162581,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.04922354,
      "balance_loss_mlp": 1.02109849,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 3.068583689958017,
      "language_loss": 0.72513795,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.74704623,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.643127202987671
    },
    {
      "auxiliary_loss_clip": 0.01122586,
      "auxiliary_loss_mlp": 0.01023751,
      "balance_loss_clip": 1.04274726,
      "balance_loss_mlp": 1.01600206,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.4885931335683473,
      "language_loss": 0.66745448,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68891782,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.7578790187835693
    },
    {
      "auxiliary_loss_clip": 0.01150496,
      "auxiliary_loss_mlp": 0.01023614,
      "balance_loss_clip": 1.04397094,
      "balance_loss_mlp": 1.01576746,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.190267269001514,
      "language_loss": 0.86563957,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88738072,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.6477434635162354
    },
    {
      "auxiliary_loss_clip": 0.01163663,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.0489099,
      "balance_loss_mlp": 1.01839817,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.8108008665593531,
      "language_loss": 0.72786891,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74976766,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.651353120803833
    },
    {
      "auxiliary_loss_clip": 0.01131569,
      "auxiliary_loss_mlp": 0.00900595,
      "balance_loss_clip": 1.04060316,
      "balance_loss_mlp": 1.00084567,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.3234365215152994,
      "language_loss": 0.7958473,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.8161689,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.8055243492126465
    },
    {
      "auxiliary_loss_clip": 0.01065264,
      "auxiliary_loss_mlp": 0.01002475,
      "balance_loss_clip": 1.0122056,
      "balance_loss_mlp": 1.00170612,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8127229437573696,
      "language_loss": 0.56178558,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58246297,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 4.166761636734009
    },
    {
      "auxiliary_loss_clip": 0.01164307,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 1.04558206,
      "balance_loss_mlp": 1.01666403,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 2.116251341741355,
      "language_loss": 0.75431812,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77620125,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.5921342372894287
    },
    {
      "auxiliary_loss_clip": 0.01062029,
      "auxiliary_loss_mlp": 0.01000965,
      "balance_loss_clip": 1.0128026,
      "balance_loss_mlp": 1.00019598,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.73044574381819,
      "language_loss": 0.64418739,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66481733,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.0860109329223633
    },
    {
      "auxiliary_loss_clip": 0.01174731,
      "auxiliary_loss_mlp": 0.01037054,
      "balance_loss_clip": 1.05118585,
      "balance_loss_mlp": 1.02937698,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 2.2455899022462087,
      "language_loss": 0.84362566,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.86574352,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.543266534805298
    },
    {
      "auxiliary_loss_clip": 0.01161266,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.04852891,
      "balance_loss_mlp": 1.01461196,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.7142226964766634,
      "language_loss": 0.81864512,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.84047651,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.6077473163604736
    },
    {
      "auxiliary_loss_clip": 0.01158584,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.04766023,
      "balance_loss_mlp": 1.01799738,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.62786880702681,
      "language_loss": 0.70003283,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.721874,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.6605148315429688
    },
    {
      "auxiliary_loss_clip": 0.01163884,
      "auxiliary_loss_mlp": 0.01029382,
      "balance_loss_clip": 1.05348682,
      "balance_loss_mlp": 1.02189243,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.4579482343690544,
      "language_loss": 0.717044,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73897666,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.709780693054199
    },
    {
      "auxiliary_loss_clip": 0.01149552,
      "auxiliary_loss_mlp": 0.01022861,
      "balance_loss_clip": 1.04552555,
      "balance_loss_mlp": 1.01523745,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 2.1617546693460494,
      "language_loss": 0.7813834,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.8031075,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.7736103534698486
    },
    {
      "auxiliary_loss_clip": 0.01169479,
      "auxiliary_loss_mlp": 0.01024308,
      "balance_loss_clip": 1.04706216,
      "balance_loss_mlp": 1.01673269,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.8892418182264585,
      "language_loss": 0.78836912,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.81030697,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.658679723739624
    },
    {
      "auxiliary_loss_clip": 0.01048628,
      "auxiliary_loss_mlp": 0.01002375,
      "balance_loss_clip": 1.01077604,
      "balance_loss_mlp": 1.00159442,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7521919946796387,
      "language_loss": 0.53829777,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55880779,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 4.199492931365967
    },
    {
      "auxiliary_loss_clip": 0.01132848,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.04739296,
      "balance_loss_mlp": 1.01796937,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.9281625257273523,
      "language_loss": 0.79379392,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81537712,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.6659128665924072
    },
    {
      "auxiliary_loss_clip": 0.01148102,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 1.04555047,
      "balance_loss_mlp": 1.01665854,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.7166350296632091,
      "language_loss": 0.71380126,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.73552728,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.6816680431365967
    },
    {
      "auxiliary_loss_clip": 0.01176136,
      "auxiliary_loss_mlp": 0.01024792,
      "balance_loss_clip": 1.04913855,
      "balance_loss_mlp": 1.01651263,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.745103281744711,
      "language_loss": 0.77477562,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79678494,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 3.5292787551879883
    },
    {
      "auxiliary_loss_clip": 0.01174942,
      "auxiliary_loss_mlp": 0.010244,
      "balance_loss_clip": 1.05048966,
      "balance_loss_mlp": 1.01647305,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.7378794396513282,
      "language_loss": 0.76668823,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.78868163,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.5549726486206055
    },
    {
      "auxiliary_loss_clip": 0.01174465,
      "auxiliary_loss_mlp": 0.01021697,
      "balance_loss_clip": 1.0511241,
      "balance_loss_mlp": 1.01407385,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 4.323288244705527,
      "language_loss": 0.77842051,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.80038214,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.546757936477661
    },
    {
      "auxiliary_loss_clip": 0.01153448,
      "auxiliary_loss_mlp": 0.01021717,
      "balance_loss_clip": 1.0456028,
      "balance_loss_mlp": 1.01413536,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.995616492380809,
      "language_loss": 0.70510775,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72685945,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.691380262374878
    },
    {
      "auxiliary_loss_clip": 0.011357,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04355288,
      "balance_loss_mlp": 1.01766181,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.1781577944530373,
      "language_loss": 0.77120131,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79281586,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.7097878456115723
    },
    {
      "auxiliary_loss_clip": 0.01162483,
      "auxiliary_loss_mlp": 0.00901227,
      "balance_loss_clip": 1.05100453,
      "balance_loss_mlp": 1.00077248,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.9927396798085917,
      "language_loss": 0.7147181,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.7353552,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.6553866863250732
    },
    {
      "auxiliary_loss_clip": 0.01142156,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.04608262,
      "balance_loss_mlp": 1.02085567,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 1.9717913947554566,
      "language_loss": 0.69904864,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72075522,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 3.7031261920928955
    },
    {
      "auxiliary_loss_clip": 0.01144792,
      "auxiliary_loss_mlp": 0.01026998,
      "balance_loss_clip": 1.04519665,
      "balance_loss_mlp": 1.0193119,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 2.180445822730485,
      "language_loss": 0.72350621,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74522412,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.707197666168213
    },
    {
      "auxiliary_loss_clip": 0.01128519,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.04301214,
      "balance_loss_mlp": 1.01994848,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 2.6301836218558794,
      "language_loss": 0.88181645,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.9033888,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.7487239837646484
    },
    {
      "auxiliary_loss_clip": 0.01156605,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 1.04494011,
      "balance_loss_mlp": 1.01783919,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.937877142490614,
      "language_loss": 0.78507805,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80690122,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.696197986602783
    },
    {
      "auxiliary_loss_clip": 0.01171195,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.05008864,
      "balance_loss_mlp": 1.01851439,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 2.1755027565347214,
      "language_loss": 0.65339279,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67536473,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.6308958530426025
    },
    {
      "auxiliary_loss_clip": 0.01156593,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.04851413,
      "balance_loss_mlp": 1.02212429,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.065020123292431,
      "language_loss": 0.70386159,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72573304,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.66308331489563
    },
    {
      "auxiliary_loss_clip": 0.01157297,
      "auxiliary_loss_mlp": 0.01027119,
      "balance_loss_clip": 1.04731917,
      "balance_loss_mlp": 1.01977539,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.6816457452346507,
      "language_loss": 0.85486507,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87670922,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.5948073863983154
    },
    {
      "auxiliary_loss_clip": 0.01152335,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.04488206,
      "balance_loss_mlp": 1.01954424,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 1.7990283577965045,
      "language_loss": 0.75766492,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.77946699,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.661518096923828
    },
    {
      "auxiliary_loss_clip": 0.01157585,
      "auxiliary_loss_mlp": 0.0102185,
      "balance_loss_clip": 1.04942369,
      "balance_loss_mlp": 1.01480508,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.7486677544125686,
      "language_loss": 0.73041546,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75220984,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 3.5482778549194336
    },
    {
      "auxiliary_loss_clip": 0.01148233,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.04551387,
      "balance_loss_mlp": 1.01675451,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.766969000058998,
      "language_loss": 0.83174998,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85347027,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.6904170513153076
    },
    {
      "auxiliary_loss_clip": 0.01164451,
      "auxiliary_loss_mlp": 0.0102523,
      "balance_loss_clip": 1.04821396,
      "balance_loss_mlp": 1.01725221,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.4243045604911635,
      "language_loss": 0.63228965,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65418655,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.789186954498291
    },
    {
      "auxiliary_loss_clip": 0.01147413,
      "auxiliary_loss_mlp": 0.01026948,
      "balance_loss_clip": 1.04595816,
      "balance_loss_mlp": 1.01962876,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 1.966213914671092,
      "language_loss": 0.747136,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76887965,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.759375810623169
    },
    {
      "auxiliary_loss_clip": 0.01166342,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04798543,
      "balance_loss_mlp": 1.01875067,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.0831757412380956,
      "language_loss": 0.72082806,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74275416,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.707662582397461
    },
    {
      "auxiliary_loss_clip": 0.01164644,
      "auxiliary_loss_mlp": 0.00901829,
      "balance_loss_clip": 1.05091071,
      "balance_loss_mlp": 1.00079298,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 2.014362764255262,
      "language_loss": 0.74150348,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.76216817,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.6018102169036865
    },
    {
      "auxiliary_loss_clip": 0.01131714,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.04244828,
      "balance_loss_mlp": 1.01948726,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.759524181927952,
      "language_loss": 0.61830848,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.63989937,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.8171491622924805
    },
    {
      "auxiliary_loss_clip": 0.01161811,
      "auxiliary_loss_mlp": 0.0102285,
      "balance_loss_clip": 1.0463779,
      "balance_loss_mlp": 1.01550996,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.8978536181484755,
      "language_loss": 0.73764294,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.75948954,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.5342307090759277
    },
    {
      "auxiliary_loss_clip": 0.01161164,
      "auxiliary_loss_mlp": 0.01023657,
      "balance_loss_clip": 1.04564905,
      "balance_loss_mlp": 1.0163759,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.5565024861959806,
      "language_loss": 0.76133245,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78318065,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.743389368057251
    },
    {
      "auxiliary_loss_clip": 0.01157761,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.04721439,
      "balance_loss_mlp": 1.01826572,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 3.2072674983403977,
      "language_loss": 0.61596918,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.63780922,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 3.6588168144226074
    },
    {
      "auxiliary_loss_clip": 0.01141448,
      "auxiliary_loss_mlp": 0.01023956,
      "balance_loss_clip": 1.04817867,
      "balance_loss_mlp": 1.01655316,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 8.49215337895358,
      "language_loss": 0.89687711,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.91853118,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.540998935699463
    },
    {
      "auxiliary_loss_clip": 0.01157335,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04576254,
      "balance_loss_mlp": 1.0169239,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 2.159889309607542,
      "language_loss": 0.60139024,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62320787,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.6553096771240234
    },
    {
      "auxiliary_loss_clip": 0.01155009,
      "auxiliary_loss_mlp": 0.01024003,
      "balance_loss_clip": 1.04587126,
      "balance_loss_mlp": 1.01740742,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 2.050515073220524,
      "language_loss": 0.82200259,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84379274,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 3.577646493911743
    },
    {
      "auxiliary_loss_clip": 0.01163003,
      "auxiliary_loss_mlp": 0.0102639,
      "balance_loss_clip": 1.0454185,
      "balance_loss_mlp": 1.01861143,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 1.8760584607414488,
      "language_loss": 0.79937732,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.8212713,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.6328063011169434
    },
    {
      "auxiliary_loss_clip": 0.01167486,
      "auxiliary_loss_mlp": 0.01020484,
      "balance_loss_clip": 1.04850626,
      "balance_loss_mlp": 1.01346278,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.4258950138971433,
      "language_loss": 0.79712236,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81900203,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.6225125789642334
    },
    {
      "auxiliary_loss_clip": 0.01157069,
      "auxiliary_loss_mlp": 0.01023277,
      "balance_loss_clip": 1.04483652,
      "balance_loss_mlp": 1.01511753,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.9279073084087721,
      "language_loss": 0.63633627,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65813977,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.625372886657715
    },
    {
      "auxiliary_loss_clip": 0.01164027,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.04660749,
      "balance_loss_mlp": 1.01955569,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.3005620850436452,
      "language_loss": 0.72350371,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74541444,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.693596601486206
    },
    {
      "auxiliary_loss_clip": 0.01158695,
      "auxiliary_loss_mlp": 0.01032374,
      "balance_loss_clip": 1.04949498,
      "balance_loss_mlp": 1.02377892,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.5615824005778727,
      "language_loss": 0.67318124,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.6950919,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.6008694171905518
    },
    {
      "auxiliary_loss_clip": 0.01161747,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04791045,
      "balance_loss_mlp": 1.01716256,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.8571835181974616,
      "language_loss": 0.73148608,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75334823,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 3.6480624675750732
    },
    {
      "auxiliary_loss_clip": 0.01139843,
      "auxiliary_loss_mlp": 0.00902215,
      "balance_loss_clip": 1.04569924,
      "balance_loss_mlp": 1.00080442,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 1.9359196782863264,
      "language_loss": 0.74596918,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76638973,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.6981139183044434
    },
    {
      "auxiliary_loss_clip": 0.01132045,
      "auxiliary_loss_mlp": 0.01022653,
      "balance_loss_clip": 1.0452404,
      "balance_loss_mlp": 1.01527941,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.8637146328704026,
      "language_loss": 0.66579437,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68734133,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.7349801063537598
    },
    {
      "auxiliary_loss_clip": 0.01153586,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.04111886,
      "balance_loss_mlp": 1.01899993,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 2.0213584424953606,
      "language_loss": 0.7486949,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77049518,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.6405298709869385
    },
    {
      "auxiliary_loss_clip": 0.01129476,
      "auxiliary_loss_mlp": 0.01026928,
      "balance_loss_clip": 1.04520631,
      "balance_loss_mlp": 1.01967978,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.5463289363605446,
      "language_loss": 0.67519325,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69675732,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.703174352645874
    },
    {
      "auxiliary_loss_clip": 0.01168251,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.04730475,
      "balance_loss_mlp": 1.01965356,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.8283575780015167,
      "language_loss": 0.72779578,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.74975324,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.627283811569214
    },
    {
      "auxiliary_loss_clip": 0.0115772,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.04540896,
      "balance_loss_mlp": 1.01846445,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.645809648581606,
      "language_loss": 0.77938068,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.80121619,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.637516975402832
    },
    {
      "auxiliary_loss_clip": 0.01130375,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.04758346,
      "balance_loss_mlp": 1.02097535,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 1.9328901341471212,
      "language_loss": 0.77924168,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.80082875,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.66520619392395
    },
    {
      "auxiliary_loss_clip": 0.01142344,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.04364634,
      "balance_loss_mlp": 1.01921725,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 2.221076292416551,
      "language_loss": 0.84240925,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86408973,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 3.597257614135742
    },
    {
      "auxiliary_loss_clip": 0.01173406,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 1.05207443,
      "balance_loss_mlp": 1.02033401,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 2.1457124228847673,
      "language_loss": 0.77343911,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79545379,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.5961227416992188
    },
    {
      "auxiliary_loss_clip": 0.01164503,
      "auxiliary_loss_mlp": 0.01024934,
      "balance_loss_clip": 1.04886854,
      "balance_loss_mlp": 1.01725721,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.668751350530516,
      "language_loss": 0.65558171,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67747605,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.53426456451416
    },
    {
      "auxiliary_loss_clip": 0.01056521,
      "auxiliary_loss_mlp": 0.01002842,
      "balance_loss_clip": 1.01558542,
      "balance_loss_mlp": 1.00187671,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7312147005447326,
      "language_loss": 0.60553867,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62613231,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.3122715950012207
    },
    {
      "auxiliary_loss_clip": 0.01157402,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.04820728,
      "balance_loss_mlp": 1.01853955,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 2.9591835796053765,
      "language_loss": 0.81651318,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83834517,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.6164307594299316
    },
    {
      "auxiliary_loss_clip": 0.0113292,
      "auxiliary_loss_mlp": 0.01023473,
      "balance_loss_clip": 1.04299474,
      "balance_loss_mlp": 1.01545298,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 1.8682650775767418,
      "language_loss": 0.75499928,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77656317,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.671464681625366
    },
    {
      "auxiliary_loss_clip": 0.01136286,
      "auxiliary_loss_mlp": 0.01022464,
      "balance_loss_clip": 1.04286075,
      "balance_loss_mlp": 1.01491237,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.835942918567078,
      "language_loss": 0.78667581,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.8082633,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.6832315921783447
    },
    {
      "auxiliary_loss_clip": 0.01149982,
      "auxiliary_loss_mlp": 0.01021497,
      "balance_loss_clip": 1.04391968,
      "balance_loss_mlp": 1.0142014,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 4.2682147896541975,
      "language_loss": 0.80156183,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82327664,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.676088809967041
    },
    {
      "auxiliary_loss_clip": 0.01157492,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05105877,
      "balance_loss_mlp": 1.0197643,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 1.8718719658816898,
      "language_loss": 0.71547222,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73731798,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.6410555839538574
    },
    {
      "auxiliary_loss_clip": 0.0116882,
      "auxiliary_loss_mlp": 0.00901707,
      "balance_loss_clip": 1.04986739,
      "balance_loss_mlp": 1.00081348,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 2.1559935637758456,
      "language_loss": 0.73948961,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.7601949,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.595587968826294
    },
    {
      "auxiliary_loss_clip": 0.01175516,
      "auxiliary_loss_mlp": 0.01029655,
      "balance_loss_clip": 1.05011547,
      "balance_loss_mlp": 1.0216502,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.252391662068859,
      "language_loss": 0.59611315,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61816484,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 3.551234722137451
    },
    {
      "auxiliary_loss_clip": 0.01129046,
      "auxiliary_loss_mlp": 0.0090107,
      "balance_loss_clip": 1.04677105,
      "balance_loss_mlp": 1.00076103,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.8729790297427542,
      "language_loss": 0.75790501,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.77820623,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.8814375400543213
    },
    {
      "auxiliary_loss_clip": 0.0104216,
      "auxiliary_loss_mlp": 0.01005056,
      "balance_loss_clip": 1.01347327,
      "balance_loss_mlp": 1.00411403,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.8457380571564183,
      "language_loss": 0.59748268,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61795479,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.3815999031066895
    },
    {
      "auxiliary_loss_clip": 0.01152184,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.04633129,
      "balance_loss_mlp": 1.02101731,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 2.1050198488400875,
      "language_loss": 0.71694106,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73875034,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 3.637913942337036
    },
    {
      "auxiliary_loss_clip": 0.01135958,
      "auxiliary_loss_mlp": 0.01029036,
      "balance_loss_clip": 1.04144514,
      "balance_loss_mlp": 1.02102506,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.9061692396139533,
      "language_loss": 0.77851337,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.80016327,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 2.6342687606811523
    },
    {
      "auxiliary_loss_clip": 0.0105708,
      "auxiliary_loss_mlp": 0.01001947,
      "balance_loss_clip": 1.01186371,
      "balance_loss_mlp": 1.00116014,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9341727739938792,
      "language_loss": 0.64962327,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.67021358,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.3097987174987793
    },
    {
      "auxiliary_loss_clip": 0.01169181,
      "auxiliary_loss_mlp": 0.01023263,
      "balance_loss_clip": 1.04686284,
      "balance_loss_mlp": 1.01655805,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.7799030139077034,
      "language_loss": 0.8031503,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.82507473,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.59000825881958
    },
    {
      "auxiliary_loss_clip": 0.01153311,
      "auxiliary_loss_mlp": 0.01024404,
      "balance_loss_clip": 1.04391837,
      "balance_loss_mlp": 1.01686108,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.7424959181268378,
      "language_loss": 0.6645298,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68630695,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.7238922119140625
    },
    {
      "auxiliary_loss_clip": 0.0116346,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 1.04947853,
      "balance_loss_mlp": 1.01574719,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 2.010415720956959,
      "language_loss": 0.7348361,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75670588,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 3.5584750175476074
    },
    {
      "auxiliary_loss_clip": 0.01141078,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.04224753,
      "balance_loss_mlp": 1.0199858,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.4616826611552236,
      "language_loss": 0.70996273,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.73165214,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.5754504203796387
    },
    {
      "auxiliary_loss_clip": 0.01171097,
      "auxiliary_loss_mlp": 0.01025734,
      "balance_loss_clip": 1.04951048,
      "balance_loss_mlp": 1.01887047,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.752627316532895,
      "language_loss": 0.72949612,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.75146449,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.5941660404205322
    },
    {
      "auxiliary_loss_clip": 0.01141038,
      "auxiliary_loss_mlp": 0.01028394,
      "balance_loss_clip": 1.04464412,
      "balance_loss_mlp": 1.02094388,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.9975804777640034,
      "language_loss": 0.73939496,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76108927,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.650228977203369
    },
    {
      "auxiliary_loss_clip": 0.01129893,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.04747367,
      "balance_loss_mlp": 1.01957583,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 3.129353243600581,
      "language_loss": 0.8356595,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85722655,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.680037260055542
    },
    {
      "auxiliary_loss_clip": 0.01142155,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 1.04653287,
      "balance_loss_mlp": 1.02348328,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 1.7457630299365672,
      "language_loss": 0.81751293,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.83925104,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.693582773208618
    },
    {
      "auxiliary_loss_clip": 0.01146322,
      "auxiliary_loss_mlp": 0.01025326,
      "balance_loss_clip": 1.04865384,
      "balance_loss_mlp": 1.01766062,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 2.0366227879087844,
      "language_loss": 0.8421998,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86391628,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.6771910190582275
    },
    {
      "auxiliary_loss_clip": 0.01126565,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 1.04119754,
      "balance_loss_mlp": 1.01880097,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 3.297062766706441,
      "language_loss": 0.65877277,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.6803025,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.692336320877075
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.04648066,
      "balance_loss_mlp": 1.01637852,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 2.2929857360014063,
      "language_loss": 0.7453112,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76719546,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 3.5222208499908447
    },
    {
      "auxiliary_loss_clip": 0.01174241,
      "auxiliary_loss_mlp": 0.01027423,
      "balance_loss_clip": 1.05378914,
      "balance_loss_mlp": 1.02004409,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.7492477591022604,
      "language_loss": 0.8768006,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.89881724,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.600332498550415
    },
    {
      "auxiliary_loss_clip": 0.01160546,
      "auxiliary_loss_mlp": 0.01030952,
      "balance_loss_clip": 1.04554343,
      "balance_loss_mlp": 1.02236259,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 1.8757961793418598,
      "language_loss": 0.73937005,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.76128501,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.6297240257263184
    },
    {
      "auxiliary_loss_clip": 0.01160294,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.04750109,
      "balance_loss_mlp": 1.01805496,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.360935150868094,
      "language_loss": 0.78592229,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.80778992,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.598334550857544
    },
    {
      "auxiliary_loss_clip": 0.01049088,
      "auxiliary_loss_mlp": 0.01002189,
      "balance_loss_clip": 1.01229715,
      "balance_loss_mlp": 1.00137806,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7895089306125801,
      "language_loss": 0.57219386,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59270662,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.1881227493286133
    },
    {
      "auxiliary_loss_clip": 0.0115345,
      "auxiliary_loss_mlp": 0.01025638,
      "balance_loss_clip": 1.04737496,
      "balance_loss_mlp": 1.01847398,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 1.961732168401316,
      "language_loss": 0.75938064,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78117156,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.682615280151367
    },
    {
      "auxiliary_loss_clip": 0.01143215,
      "auxiliary_loss_mlp": 0.01023206,
      "balance_loss_clip": 1.04471505,
      "balance_loss_mlp": 1.0161221,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.9982322557707763,
      "language_loss": 0.75641215,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.77807635,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.712301254272461
    },
    {
      "auxiliary_loss_clip": 0.01140776,
      "auxiliary_loss_mlp": 0.01023704,
      "balance_loss_clip": 1.04219842,
      "balance_loss_mlp": 1.01675165,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.1438906635886377,
      "language_loss": 0.85128808,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87293285,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.7500600814819336
    },
    {
      "auxiliary_loss_clip": 0.01161851,
      "auxiliary_loss_mlp": 0.01025368,
      "balance_loss_clip": 1.04808378,
      "balance_loss_mlp": 1.01859713,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 1.8949996201300536,
      "language_loss": 0.81048429,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83235645,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.7642629146575928
    },
    {
      "auxiliary_loss_clip": 0.0115531,
      "auxiliary_loss_mlp": 0.01026583,
      "balance_loss_clip": 1.0475992,
      "balance_loss_mlp": 1.01899505,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 3.0063870969104283,
      "language_loss": 0.76922703,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.7910459,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.5959017276763916
    },
    {
      "auxiliary_loss_clip": 0.01156157,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.04664242,
      "balance_loss_mlp": 1.01546454,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 2.537924465377254,
      "language_loss": 0.74261427,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76440316,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 3.533930778503418
    },
    {
      "auxiliary_loss_clip": 0.01168606,
      "auxiliary_loss_mlp": 0.01030112,
      "balance_loss_clip": 1.05009174,
      "balance_loss_mlp": 1.02222645,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.3548314272381514,
      "language_loss": 0.72727537,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74926251,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.580537796020508
    },
    {
      "auxiliary_loss_clip": 0.01128005,
      "auxiliary_loss_mlp": 0.01028697,
      "balance_loss_clip": 1.04434252,
      "balance_loss_mlp": 1.02116597,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 1.989945034954662,
      "language_loss": 0.74991119,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.7714783,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.7268168926239014
    },
    {
      "auxiliary_loss_clip": 0.01155532,
      "auxiliary_loss_mlp": 0.01020944,
      "balance_loss_clip": 1.04378259,
      "balance_loss_mlp": 1.01374388,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 2.4632422036365447,
      "language_loss": 0.82172179,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84348655,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.6287970542907715
    },
    {
      "auxiliary_loss_clip": 0.01160329,
      "auxiliary_loss_mlp": 0.01021442,
      "balance_loss_clip": 1.04687464,
      "balance_loss_mlp": 1.01465607,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.3599031058088022,
      "language_loss": 0.76617348,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78799123,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 3.448528528213501
    },
    {
      "auxiliary_loss_clip": 0.0115332,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.04640818,
      "balance_loss_mlp": 1.01822472,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 2.3877181923206323,
      "language_loss": 0.67380482,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69559681,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.5937728881835938
    },
    {
      "auxiliary_loss_clip": 0.01134228,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.04643309,
      "balance_loss_mlp": 1.02359045,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.169873183271477,
      "language_loss": 0.6974895,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.71914589,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.6802923679351807
    },
    {
      "auxiliary_loss_clip": 0.01156102,
      "auxiliary_loss_mlp": 0.01022103,
      "balance_loss_clip": 1.04748535,
      "balance_loss_mlp": 1.01444387,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.6482372658787996,
      "language_loss": 0.56943166,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.5912137,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.689833879470825
    },
    {
      "auxiliary_loss_clip": 0.01150119,
      "auxiliary_loss_mlp": 0.01024872,
      "balance_loss_clip": 1.04618645,
      "balance_loss_mlp": 1.01714706,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.4332095317524303,
      "language_loss": 0.75855285,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.7803027,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.6180388927459717
    },
    {
      "auxiliary_loss_clip": 0.01128607,
      "auxiliary_loss_mlp": 0.01022816,
      "balance_loss_clip": 1.03823972,
      "balance_loss_mlp": 1.01511502,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.5471308229806027,
      "language_loss": 0.64096189,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66247612,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.960278272628784
    },
    {
      "auxiliary_loss_clip": 0.0115688,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.04740191,
      "balance_loss_mlp": 1.02024078,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.1870870951751664,
      "language_loss": 0.53575408,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55760276,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.6389992237091064
    },
    {
      "auxiliary_loss_clip": 0.01171787,
      "auxiliary_loss_mlp": 0.0102347,
      "balance_loss_clip": 1.04998565,
      "balance_loss_mlp": 1.01642203,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 4.73452918007852,
      "language_loss": 0.83993775,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.86189032,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.537381410598755
    },
    {
      "auxiliary_loss_clip": 0.01147609,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 1.04387653,
      "balance_loss_mlp": 1.01828909,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 1.915788358712734,
      "language_loss": 0.75670773,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77844626,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.641852378845215
    },
    {
      "auxiliary_loss_clip": 0.01156927,
      "auxiliary_loss_mlp": 0.01025631,
      "balance_loss_clip": 1.04832196,
      "balance_loss_mlp": 1.01843691,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.233622236899392,
      "language_loss": 0.77980787,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80163348,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.5900046825408936
    },
    {
      "auxiliary_loss_clip": 0.01171773,
      "auxiliary_loss_mlp": 0.01023993,
      "balance_loss_clip": 1.04736876,
      "balance_loss_mlp": 1.01646543,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.296941452256506,
      "language_loss": 0.79457533,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81653297,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.5156846046447754
    },
    {
      "auxiliary_loss_clip": 0.01124482,
      "auxiliary_loss_mlp": 0.00901276,
      "balance_loss_clip": 1.04117203,
      "balance_loss_mlp": 1.00083971,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 1.9976446563698622,
      "language_loss": 0.65268481,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.6729424,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.762465715408325
    },
    {
      "auxiliary_loss_clip": 0.01161315,
      "auxiliary_loss_mlp": 0.01025643,
      "balance_loss_clip": 1.04844761,
      "balance_loss_mlp": 1.01812053,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 2.321553780461255,
      "language_loss": 0.77497286,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79684246,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.5297799110412598
    },
    {
      "auxiliary_loss_clip": 0.011644,
      "auxiliary_loss_mlp": 0.01021498,
      "balance_loss_clip": 1.04688287,
      "balance_loss_mlp": 1.01440835,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.829300871880274,
      "language_loss": 0.78991103,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81176996,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 3.501577615737915
    },
    {
      "auxiliary_loss_clip": 0.01134607,
      "auxiliary_loss_mlp": 0.01023336,
      "balance_loss_clip": 1.04192173,
      "balance_loss_mlp": 1.01591587,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.6307108794817993,
      "language_loss": 0.82795066,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.84953016,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.671262741088867
    },
    {
      "auxiliary_loss_clip": 0.01149399,
      "auxiliary_loss_mlp": 0.01022283,
      "balance_loss_clip": 1.04514873,
      "balance_loss_mlp": 1.01533318,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 2.029391695397201,
      "language_loss": 0.7817781,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80349493,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.6350810527801514
    },
    {
      "auxiliary_loss_clip": 0.01159521,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 1.04752386,
      "balance_loss_mlp": 1.01688552,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 2.176592166329112,
      "language_loss": 0.69101077,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71284842,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.6093881130218506
    },
    {
      "auxiliary_loss_clip": 0.01165628,
      "auxiliary_loss_mlp": 0.01024344,
      "balance_loss_clip": 1.04444313,
      "balance_loss_mlp": 1.01607704,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.7642034177451744,
      "language_loss": 0.77618206,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79808176,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.669182538986206
    },
    {
      "auxiliary_loss_clip": 0.01151656,
      "auxiliary_loss_mlp": 0.01024911,
      "balance_loss_clip": 1.04519653,
      "balance_loss_mlp": 1.01756144,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 2.0194599650427016,
      "language_loss": 0.80391806,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82568371,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.6068203449249268
    },
    {
      "auxiliary_loss_clip": 0.0116932,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.04567683,
      "balance_loss_mlp": 1.01825809,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 1.7510387360907986,
      "language_loss": 0.75436252,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77631807,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.5477190017700195
    },
    {
      "auxiliary_loss_clip": 0.01161797,
      "auxiliary_loss_mlp": 0.01032889,
      "balance_loss_clip": 1.04971325,
      "balance_loss_mlp": 1.02517653,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 3.159123255176842,
      "language_loss": 0.76622903,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78817588,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.595162868499756
    },
    {
      "auxiliary_loss_clip": 0.01159168,
      "auxiliary_loss_mlp": 0.01022823,
      "balance_loss_clip": 1.04703796,
      "balance_loss_mlp": 1.01554513,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.4230537081344565,
      "language_loss": 0.82334292,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84516287,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.60016131401062
    },
    {
      "auxiliary_loss_clip": 0.01147966,
      "auxiliary_loss_mlp": 0.00901529,
      "balance_loss_clip": 1.04257643,
      "balance_loss_mlp": 1.00080371,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.84995947285202,
      "language_loss": 0.71765733,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.73815227,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.6653645038604736
    },
    {
      "auxiliary_loss_clip": 0.01147928,
      "auxiliary_loss_mlp": 0.01024227,
      "balance_loss_clip": 1.04578638,
      "balance_loss_mlp": 1.01699746,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 3.4517397160911085,
      "language_loss": 0.83277345,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85449499,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 3.5929453372955322
    },
    {
      "auxiliary_loss_clip": 0.01066301,
      "auxiliary_loss_mlp": 0.01001868,
      "balance_loss_clip": 1.01153135,
      "balance_loss_mlp": 1.00106299,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8154590654653657,
      "language_loss": 0.59505928,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.6157409,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.1585466861724854
    },
    {
      "auxiliary_loss_clip": 0.011485,
      "auxiliary_loss_mlp": 0.01024792,
      "balance_loss_clip": 1.04348922,
      "balance_loss_mlp": 1.01714444,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 3.1573839838994697,
      "language_loss": 0.70127159,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72300446,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.711153030395508
    },
    {
      "auxiliary_loss_clip": 0.0115419,
      "auxiliary_loss_mlp": 0.00901358,
      "balance_loss_clip": 1.04954827,
      "balance_loss_mlp": 1.00073195,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.682431544186347,
      "language_loss": 0.81899756,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.839553,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 3.559971809387207
    },
    {
      "auxiliary_loss_clip": 0.01154844,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.04618168,
      "balance_loss_mlp": 1.02175879,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.0029936731068587,
      "language_loss": 0.85901117,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88085341,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.5502560138702393
    },
    {
      "auxiliary_loss_clip": 0.01153347,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.04719949,
      "balance_loss_mlp": 1.01769125,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.2958314730749407,
      "language_loss": 0.69781184,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71959698,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 2.628570556640625
    },
    {
      "auxiliary_loss_clip": 0.01165679,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.04567397,
      "balance_loss_mlp": 1.0194664,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.0996114347872896,
      "language_loss": 0.76483715,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78675705,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.4957590103149414
    },
    {
      "auxiliary_loss_clip": 0.0116607,
      "auxiliary_loss_mlp": 0.01023863,
      "balance_loss_clip": 1.04683518,
      "balance_loss_mlp": 1.0161531,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.5043179326568508,
      "language_loss": 0.81567633,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83757567,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.6878867149353027
    },
    {
      "auxiliary_loss_clip": 0.01152163,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.04460061,
      "balance_loss_mlp": 1.01950228,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 2.3148922118080937,
      "language_loss": 0.67708874,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.69888788,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.7854440212249756
    },
    {
      "auxiliary_loss_clip": 0.01159031,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.04676223,
      "balance_loss_mlp": 1.0192616,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 2.101526641450026,
      "language_loss": 0.66975081,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.69160295,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 3.5294477939605713
    },
    {
      "auxiliary_loss_clip": 0.01163284,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.04872513,
      "balance_loss_mlp": 1.01815712,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 2.0351139275396695,
      "language_loss": 0.73845339,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.76033568,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.609222888946533
    },
    {
      "auxiliary_loss_clip": 0.01159135,
      "auxiliary_loss_mlp": 0.01032574,
      "balance_loss_clip": 1.04734373,
      "balance_loss_mlp": 1.02457547,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 2.1813274651725654,
      "language_loss": 0.75291169,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77482879,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.6553094387054443
    },
    {
      "auxiliary_loss_clip": 0.01139907,
      "auxiliary_loss_mlp": 0.00900717,
      "balance_loss_clip": 1.04229879,
      "balance_loss_mlp": 1.00073647,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 2.6190944611987916,
      "language_loss": 0.85976934,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.88017559,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.6622848510742188
    },
    {
      "auxiliary_loss_clip": 0.01163982,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.04560494,
      "balance_loss_mlp": 1.01725435,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.779695689266683,
      "language_loss": 0.80894268,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.83082879,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.5551552772521973
    },
    {
      "auxiliary_loss_clip": 0.01162331,
      "auxiliary_loss_mlp": 0.01021309,
      "balance_loss_clip": 1.04861522,
      "balance_loss_mlp": 1.01379251,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.9583815356890695,
      "language_loss": 0.73700613,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75884253,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.508553981781006
    },
    {
      "auxiliary_loss_clip": 0.01170518,
      "auxiliary_loss_mlp": 0.01029798,
      "balance_loss_clip": 1.05003166,
      "balance_loss_mlp": 1.02146542,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.9097833570002374,
      "language_loss": 0.72263372,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74463689,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.5314600467681885
    },
    {
      "auxiliary_loss_clip": 0.01062285,
      "auxiliary_loss_mlp": 0.01000623,
      "balance_loss_clip": 1.01169062,
      "balance_loss_mlp": 0.99984831,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7594614015227121,
      "language_loss": 0.5734331,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59406221,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.1211507320404053
    },
    {
      "auxiliary_loss_clip": 0.01176512,
      "auxiliary_loss_mlp": 0.01022587,
      "balance_loss_clip": 1.05226743,
      "balance_loss_mlp": 1.01512742,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.436159609323451,
      "language_loss": 0.79131031,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81330132,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 3.7690510749816895
    },
    {
      "auxiliary_loss_clip": 0.01148868,
      "auxiliary_loss_mlp": 0.01030394,
      "balance_loss_clip": 1.04670429,
      "balance_loss_mlp": 1.02225161,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 2.208688669882258,
      "language_loss": 0.85598755,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.8777802,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.6486244201660156
    },
    {
      "auxiliary_loss_clip": 0.01169774,
      "auxiliary_loss_mlp": 0.01026118,
      "balance_loss_clip": 1.04707837,
      "balance_loss_mlp": 1.01858079,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.8667040944818534,
      "language_loss": 0.73866314,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.76062202,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.5039515495300293
    },
    {
      "auxiliary_loss_clip": 0.01127939,
      "auxiliary_loss_mlp": 0.00900935,
      "balance_loss_clip": 1.03880811,
      "balance_loss_mlp": 1.0007621,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 2.494779601937878,
      "language_loss": 0.82399023,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.84427905,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.739900827407837
    },
    {
      "auxiliary_loss_clip": 0.01164312,
      "auxiliary_loss_mlp": 0.01023948,
      "balance_loss_clip": 1.04842114,
      "balance_loss_mlp": 1.01644373,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 2.0570135803207825,
      "language_loss": 0.82686257,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.84874517,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.6579031944274902
    },
    {
      "auxiliary_loss_clip": 0.01172405,
      "auxiliary_loss_mlp": 0.0090155,
      "balance_loss_clip": 1.04880333,
      "balance_loss_mlp": 1.00072694,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 3.7616893938813014,
      "language_loss": 0.65767127,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67841083,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.573969841003418
    },
    {
      "auxiliary_loss_clip": 0.01166764,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.04588175,
      "balance_loss_mlp": 1.01772046,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 2.6249713217030344,
      "language_loss": 0.63137496,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.6533016,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.575413703918457
    },
    {
      "auxiliary_loss_clip": 0.01071808,
      "auxiliary_loss_mlp": 0.01001181,
      "balance_loss_clip": 1.01178277,
      "balance_loss_mlp": 1.00034702,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7126998390652962,
      "language_loss": 0.57469058,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59542042,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.2248077392578125
    },
    {
      "auxiliary_loss_clip": 0.01178046,
      "auxiliary_loss_mlp": 0.01024889,
      "balance_loss_clip": 1.05031443,
      "balance_loss_mlp": 1.01623988,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 2.1068800704213575,
      "language_loss": 0.70184994,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72387934,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.672567129135132
    },
    {
      "auxiliary_loss_clip": 0.01147152,
      "auxiliary_loss_mlp": 0.01022789,
      "balance_loss_clip": 1.04417944,
      "balance_loss_mlp": 1.01530564,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 2.030679096638714,
      "language_loss": 0.74528551,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.76698494,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 3.5419905185699463
    },
    {
      "auxiliary_loss_clip": 0.01169672,
      "auxiliary_loss_mlp": 0.01023937,
      "balance_loss_clip": 1.04825473,
      "balance_loss_mlp": 1.01682007,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.8144876601628475,
      "language_loss": 0.81563133,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83756739,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.6847543716430664
    },
    {
      "auxiliary_loss_clip": 0.01165529,
      "auxiliary_loss_mlp": 0.01032758,
      "balance_loss_clip": 1.05060029,
      "balance_loss_mlp": 1.02572763,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.6273344405281964,
      "language_loss": 0.71045351,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73243642,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 2.8920979499816895
    },
    {
      "auxiliary_loss_clip": 0.0104178,
      "auxiliary_loss_mlp": 0.0100391,
      "balance_loss_clip": 1.01194632,
      "balance_loss_mlp": 1.0030396,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8410904734518153,
      "language_loss": 0.50358784,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52404475,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.326245069503784
    },
    {
      "auxiliary_loss_clip": 0.01148574,
      "auxiliary_loss_mlp": 0.01029714,
      "balance_loss_clip": 1.04920936,
      "balance_loss_mlp": 1.0227133,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 2.5476576529853774,
      "language_loss": 0.71607399,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73785686,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 3.796531915664673
    },
    {
      "auxiliary_loss_clip": 0.01140932,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.04560268,
      "balance_loss_mlp": 1.02030754,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 1.9797037884466722,
      "language_loss": 0.77535397,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.79703975,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.718733072280884
    },
    {
      "auxiliary_loss_clip": 0.01145503,
      "auxiliary_loss_mlp": 0.01022326,
      "balance_loss_clip": 1.04686546,
      "balance_loss_mlp": 1.01430929,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 2.302294849692401,
      "language_loss": 0.65405834,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67573661,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.6807944774627686
    },
    {
      "auxiliary_loss_clip": 0.01059971,
      "auxiliary_loss_mlp": 0.01002262,
      "balance_loss_clip": 1.01047754,
      "balance_loss_mlp": 1.00143945,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8478521264829295,
      "language_loss": 0.6400218,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66064417,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.11244535446167
    },
    {
      "auxiliary_loss_clip": 0.01131264,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.03684115,
      "balance_loss_mlp": 1.01848102,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.681173724137582,
      "language_loss": 0.73964012,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76120746,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.7364635467529297
    },
    {
      "auxiliary_loss_clip": 0.01159349,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 1.04940653,
      "balance_loss_mlp": 1.0209167,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.1445943023674587,
      "language_loss": 0.82858324,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.85045958,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 3.4892218112945557
    },
    {
      "auxiliary_loss_clip": 0.01168694,
      "auxiliary_loss_mlp": 0.01020919,
      "balance_loss_clip": 1.05109334,
      "balance_loss_mlp": 1.01364076,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.4945121646066366,
      "language_loss": 0.79618174,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.81807786,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.6054654121398926
    },
    {
      "auxiliary_loss_clip": 0.0115039,
      "auxiliary_loss_mlp": 0.01022048,
      "balance_loss_clip": 1.04692304,
      "balance_loss_mlp": 1.01469278,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 2.4441981228637,
      "language_loss": 0.73873466,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.76045907,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.640063524246216
    },
    {
      "auxiliary_loss_clip": 0.01153112,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.04542315,
      "balance_loss_mlp": 1.01977372,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.3387467684213603,
      "language_loss": 0.71688414,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.73869306,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.620422124862671
    },
    {
      "auxiliary_loss_clip": 0.01127049,
      "auxiliary_loss_mlp": 0.01018843,
      "balance_loss_clip": 1.04017568,
      "balance_loss_mlp": 1.01150274,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.934163769229271,
      "language_loss": 0.80191296,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82337189,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.8187057971954346
    },
    {
      "auxiliary_loss_clip": 0.01144966,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.0407995,
      "balance_loss_mlp": 1.02242923,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.24795563585865,
      "language_loss": 0.66239208,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68414724,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.761815071105957
    },
    {
      "auxiliary_loss_clip": 0.01164792,
      "auxiliary_loss_mlp": 0.00901242,
      "balance_loss_clip": 1.04803252,
      "balance_loss_mlp": 1.0008862,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.2583752689998238,
      "language_loss": 0.91215128,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.93281162,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.546647071838379
    },
    {
      "auxiliary_loss_clip": 0.01054309,
      "auxiliary_loss_mlp": 0.010013,
      "balance_loss_clip": 1.01395476,
      "balance_loss_mlp": 1.00055242,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8863906387879478,
      "language_loss": 0.6306724,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65122849,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.2608578205108643
    },
    {
      "auxiliary_loss_clip": 0.01161862,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.04702532,
      "balance_loss_mlp": 1.02081609,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 1.8482971959625896,
      "language_loss": 0.80699563,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.82889915,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 3.4692277908325195
    },
    {
      "auxiliary_loss_clip": 0.01151789,
      "auxiliary_loss_mlp": 0.01023948,
      "balance_loss_clip": 1.04892075,
      "balance_loss_mlp": 1.01630688,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.0989255636846464,
      "language_loss": 0.66283381,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68459117,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.6325595378875732
    },
    {
      "auxiliary_loss_clip": 0.01164883,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.04887426,
      "balance_loss_mlp": 1.0185585,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.7366100414028331,
      "language_loss": 0.71304315,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73495525,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.5730531215667725
    },
    {
      "auxiliary_loss_clip": 0.01143706,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.0435816,
      "balance_loss_mlp": 1.01977754,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 2.091209746839851,
      "language_loss": 0.71203589,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73374683,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.7463972568511963
    },
    {
      "auxiliary_loss_clip": 0.01146966,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 1.0455544,
      "balance_loss_mlp": 1.02011383,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 2.377768776312558,
      "language_loss": 0.7218492,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.7435931,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.7973134517669678
    },
    {
      "auxiliary_loss_clip": 0.01132607,
      "auxiliary_loss_mlp": 0.01033122,
      "balance_loss_clip": 1.04505658,
      "balance_loss_mlp": 1.02479517,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.4120223715484905,
      "language_loss": 0.84637725,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.8680346,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.713658571243286
    },
    {
      "auxiliary_loss_clip": 0.01164439,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.0473001,
      "balance_loss_mlp": 1.02068067,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 1.8468509254748278,
      "language_loss": 0.77312618,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79505384,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.6724202632904053
    },
    {
      "auxiliary_loss_clip": 0.01148046,
      "auxiliary_loss_mlp": 0.01025945,
      "balance_loss_clip": 1.04388571,
      "balance_loss_mlp": 1.01805353,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.192729314397734,
      "language_loss": 0.72975802,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.75149798,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.5845611095428467
    },
    {
      "auxiliary_loss_clip": 0.01177017,
      "auxiliary_loss_mlp": 0.01027446,
      "balance_loss_clip": 1.05298996,
      "balance_loss_mlp": 1.01998949,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 2.025732952238667,
      "language_loss": 0.68616718,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70821178,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.7260825634002686
    },
    {
      "auxiliary_loss_clip": 0.01162066,
      "auxiliary_loss_mlp": 0.01023637,
      "balance_loss_clip": 1.04773712,
      "balance_loss_mlp": 1.01633525,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 1.9934277526947501,
      "language_loss": 0.81581116,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83766818,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 3.52392578125
    },
    {
      "auxiliary_loss_clip": 0.01141927,
      "auxiliary_loss_mlp": 0.01023697,
      "balance_loss_clip": 1.04349005,
      "balance_loss_mlp": 1.01679754,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.9673497020924418,
      "language_loss": 0.80413336,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82578957,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.726656436920166
    },
    {
      "auxiliary_loss_clip": 0.01169912,
      "auxiliary_loss_mlp": 0.01022463,
      "balance_loss_clip": 1.04786682,
      "balance_loss_mlp": 1.01541221,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.6213443062148822,
      "language_loss": 0.79380792,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81573164,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.554152250289917
    },
    {
      "auxiliary_loss_clip": 0.01170278,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.048455,
      "balance_loss_mlp": 1.01842833,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.9104529203579033,
      "language_loss": 0.73391277,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75586802,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.635838031768799
    },
    {
      "auxiliary_loss_clip": 0.01166898,
      "auxiliary_loss_mlp": 0.0103002,
      "balance_loss_clip": 1.04751444,
      "balance_loss_mlp": 1.02268827,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 3.8897241299703604,
      "language_loss": 0.72336996,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74533916,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.5633864402770996
    },
    {
      "auxiliary_loss_clip": 0.01123517,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.0370971,
      "balance_loss_mlp": 1.01711464,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.7772023379321755,
      "language_loss": 0.59224409,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61372936,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 3.6688761711120605
    },
    {
      "auxiliary_loss_clip": 0.0114699,
      "auxiliary_loss_mlp": 0.01029799,
      "balance_loss_clip": 1.04505587,
      "balance_loss_mlp": 1.02238381,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 3.6004003840004324,
      "language_loss": 0.86660904,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88837689,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.600850820541382
    },
    {
      "auxiliary_loss_clip": 0.01167828,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.05135489,
      "balance_loss_mlp": 1.01678598,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.8368236918059142,
      "language_loss": 0.81783581,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83975279,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.547123432159424
    },
    {
      "auxiliary_loss_clip": 0.01154046,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.04131579,
      "balance_loss_mlp": 1.01867747,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.7796800129162516,
      "language_loss": 0.75542986,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.77723575,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.557018756866455
    },
    {
      "auxiliary_loss_clip": 0.01152422,
      "auxiliary_loss_mlp": 0.01022473,
      "balance_loss_clip": 1.04778719,
      "balance_loss_mlp": 1.01461673,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.372771352288107,
      "language_loss": 0.65028179,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67203081,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.5878851413726807
    },
    {
      "auxiliary_loss_clip": 0.01155599,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.04945421,
      "balance_loss_mlp": 1.01600718,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 1.909537160195459,
      "language_loss": 0.83729935,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.85908037,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 3.465693473815918
    },
    {
      "auxiliary_loss_clip": 0.01148902,
      "auxiliary_loss_mlp": 0.01021498,
      "balance_loss_clip": 1.04586208,
      "balance_loss_mlp": 1.01392221,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 5.8144487158404585,
      "language_loss": 0.74240184,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76410586,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.701420783996582
    },
    {
      "auxiliary_loss_clip": 0.0115207,
      "auxiliary_loss_mlp": 0.01025343,
      "balance_loss_clip": 1.04523396,
      "balance_loss_mlp": 1.01844096,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.5844603553125043,
      "language_loss": 0.73321444,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.75498855,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.619798183441162
    },
    {
      "auxiliary_loss_clip": 0.01067839,
      "auxiliary_loss_mlp": 0.01000632,
      "balance_loss_clip": 1.01149464,
      "balance_loss_mlp": 0.99980974,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.810738400984182,
      "language_loss": 0.57610685,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59679157,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.0835413932800293
    },
    {
      "auxiliary_loss_clip": 0.01143149,
      "auxiliary_loss_mlp": 0.0102376,
      "balance_loss_clip": 1.04127145,
      "balance_loss_mlp": 1.01723933,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 1.9956109946134872,
      "language_loss": 0.83335245,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85502154,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.6753647327423096
    },
    {
      "auxiliary_loss_clip": 0.01161972,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.04838252,
      "balance_loss_mlp": 1.02153754,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 3.5805324471301816,
      "language_loss": 0.76002312,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78193337,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.6922731399536133
    },
    {
      "auxiliary_loss_clip": 0.01162813,
      "auxiliary_loss_mlp": 0.01023066,
      "balance_loss_clip": 1.04890728,
      "balance_loss_mlp": 1.01571071,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.5043520454197852,
      "language_loss": 0.81160688,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.8334657,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.6073617935180664
    },
    {
      "auxiliary_loss_clip": 0.01136458,
      "auxiliary_loss_mlp": 0.01024648,
      "balance_loss_clip": 1.0462352,
      "balance_loss_mlp": 1.0173583,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 3.2707262400007098,
      "language_loss": 0.73664749,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75825852,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.805194854736328
    },
    {
      "auxiliary_loss_clip": 0.01167081,
      "auxiliary_loss_mlp": 0.01020995,
      "balance_loss_clip": 1.0473274,
      "balance_loss_mlp": 1.0133481,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.6246047273443278,
      "language_loss": 0.71557623,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73745698,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 3.5897035598754883
    },
    {
      "auxiliary_loss_clip": 0.01151863,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.04474831,
      "balance_loss_mlp": 1.01713777,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.799621170744335,
      "language_loss": 0.75802159,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.77978504,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 2.650522470474243
    },
    {
      "auxiliary_loss_clip": 0.01148654,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.04501176,
      "balance_loss_mlp": 1.01859641,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 3.2987636987147226,
      "language_loss": 0.77977461,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80152428,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.650820016860962
    },
    {
      "auxiliary_loss_clip": 0.01165048,
      "auxiliary_loss_mlp": 0.01026588,
      "balance_loss_clip": 1.05270314,
      "balance_loss_mlp": 1.01887465,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.8293832083980712,
      "language_loss": 0.69121742,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71313381,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.8166286945343018
    },
    {
      "auxiliary_loss_clip": 0.01165881,
      "auxiliary_loss_mlp": 0.00901454,
      "balance_loss_clip": 1.0500288,
      "balance_loss_mlp": 1.00072658,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 2.741405847734064,
      "language_loss": 0.76520032,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78587365,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.6717212200164795
    },
    {
      "auxiliary_loss_clip": 0.01121224,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.04233742,
      "balance_loss_mlp": 1.01662874,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.1941600409059996,
      "language_loss": 0.74785686,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.76931196,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.746600866317749
    },
    {
      "auxiliary_loss_clip": 0.01154406,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.04713917,
      "balance_loss_mlp": 1.02086186,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.145294746455669,
      "language_loss": 0.65146351,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67328918,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.6125130653381348
    },
    {
      "auxiliary_loss_clip": 0.01167001,
      "auxiliary_loss_mlp": 0.01021653,
      "balance_loss_clip": 1.04774916,
      "balance_loss_mlp": 1.01423228,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 2.0691206100895525,
      "language_loss": 0.69101006,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71289659,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.588465452194214
    },
    {
      "auxiliary_loss_clip": 0.01152845,
      "auxiliary_loss_mlp": 0.00900855,
      "balance_loss_clip": 1.04359865,
      "balance_loss_mlp": 1.00083089,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.6539778307077726,
      "language_loss": 0.7863999,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80693686,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.638474225997925
    },
    {
      "auxiliary_loss_clip": 0.01167589,
      "auxiliary_loss_mlp": 0.01024604,
      "balance_loss_clip": 1.04644847,
      "balance_loss_mlp": 1.01687956,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 1.9945299152795217,
      "language_loss": 0.77035165,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79227358,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 3.565308094024658
    },
    {
      "auxiliary_loss_clip": 0.01137602,
      "auxiliary_loss_mlp": 0.01026348,
      "balance_loss_clip": 1.04395986,
      "balance_loss_mlp": 1.01883757,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.063792866355453,
      "language_loss": 0.78079873,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.8024382,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.714118480682373
    },
    {
      "auxiliary_loss_clip": 0.01174062,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.0516336,
      "balance_loss_mlp": 1.01810336,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 2.103695465390924,
      "language_loss": 0.83578855,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85778069,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.556713104248047
    },
    {
      "auxiliary_loss_clip": 0.01154411,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04570317,
      "balance_loss_mlp": 1.01855373,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.1905186814822732,
      "language_loss": 0.78907955,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.81088275,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.7104032039642334
    },
    {
      "auxiliary_loss_clip": 0.01162849,
      "auxiliary_loss_mlp": 0.01019878,
      "balance_loss_clip": 1.04555011,
      "balance_loss_mlp": 1.0123086,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 2.9869972940147833,
      "language_loss": 0.67961925,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70144653,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.657296657562256
    },
    {
      "auxiliary_loss_clip": 0.01155594,
      "auxiliary_loss_mlp": 0.01023777,
      "balance_loss_clip": 1.04569626,
      "balance_loss_mlp": 1.01625466,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 2.221044352387612,
      "language_loss": 0.78502023,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80681396,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 3.6424450874328613
    },
    {
      "auxiliary_loss_clip": 0.01141072,
      "auxiliary_loss_mlp": 0.01026681,
      "balance_loss_clip": 1.0462389,
      "balance_loss_mlp": 1.02001739,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 2.3825008997042127,
      "language_loss": 0.74565768,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.76733524,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.782750368118286
    },
    {
      "auxiliary_loss_clip": 0.01165608,
      "auxiliary_loss_mlp": 0.01027568,
      "balance_loss_clip": 1.04795563,
      "balance_loss_mlp": 1.02029884,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.868573661408567,
      "language_loss": 0.68956119,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71149296,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.7904276847839355
    },
    {
      "auxiliary_loss_clip": 0.0114472,
      "auxiliary_loss_mlp": 0.01024243,
      "balance_loss_clip": 1.04373503,
      "balance_loss_mlp": 1.01714754,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 2.0500351982765794,
      "language_loss": 0.79932165,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.8210113,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.718200922012329
    },
    {
      "auxiliary_loss_clip": 0.01153081,
      "auxiliary_loss_mlp": 0.01023366,
      "balance_loss_clip": 1.04564524,
      "balance_loss_mlp": 1.01588523,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.9518252847996658,
      "language_loss": 0.79865032,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.82041478,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 3.66849684715271
    },
    {
      "auxiliary_loss_clip": 0.01147378,
      "auxiliary_loss_mlp": 0.01023289,
      "balance_loss_clip": 1.04560065,
      "balance_loss_mlp": 1.01646137,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 2.6852366544079476,
      "language_loss": 0.84639013,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86809677,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.677985429763794
    },
    {
      "auxiliary_loss_clip": 0.01156492,
      "auxiliary_loss_mlp": 0.01020822,
      "balance_loss_clip": 1.04454041,
      "balance_loss_mlp": 1.01393223,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.7267502599070261,
      "language_loss": 0.66823667,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.69000977,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.616119384765625
    },
    {
      "auxiliary_loss_clip": 0.01154252,
      "auxiliary_loss_mlp": 0.01027023,
      "balance_loss_clip": 1.04981065,
      "balance_loss_mlp": 1.0196259,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.6814490338581842,
      "language_loss": 0.80819976,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.83001244,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.76716685295105
    },
    {
      "auxiliary_loss_clip": 0.01132362,
      "auxiliary_loss_mlp": 0.01022318,
      "balance_loss_clip": 1.04431236,
      "balance_loss_mlp": 1.01520467,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 2.024687024162419,
      "language_loss": 0.70339686,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72494364,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.6426451206207275
    },
    {
      "auxiliary_loss_clip": 0.01169975,
      "auxiliary_loss_mlp": 0.01027853,
      "balance_loss_clip": 1.04754782,
      "balance_loss_mlp": 1.02013993,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 2.1620630467081154,
      "language_loss": 0.73822761,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.76020586,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.620776891708374
    },
    {
      "auxiliary_loss_clip": 0.0116696,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.05034912,
      "balance_loss_mlp": 1.01663244,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 1.8049817091090077,
      "language_loss": 0.81016695,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83208483,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.5825419425964355
    },
    {
      "auxiliary_loss_clip": 0.01158305,
      "auxiliary_loss_mlp": 0.01029752,
      "balance_loss_clip": 1.0472672,
      "balance_loss_mlp": 1.02244413,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 2.587527579007068,
      "language_loss": 0.74067569,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.7625562,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.641676664352417
    },
    {
      "auxiliary_loss_clip": 0.01161175,
      "auxiliary_loss_mlp": 0.01023804,
      "balance_loss_clip": 1.04650795,
      "balance_loss_mlp": 1.01673222,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 2.8801568466750114,
      "language_loss": 0.7558161,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77766591,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 3.591831922531128
    },
    {
      "auxiliary_loss_clip": 0.01143152,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.04427314,
      "balance_loss_mlp": 1.01733541,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.8750155943866857,
      "language_loss": 0.72415745,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74583912,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.8980748653411865
    },
    {
      "auxiliary_loss_clip": 0.01161553,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.04823768,
      "balance_loss_mlp": 1.02069962,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.876083071697533,
      "language_loss": 0.7082482,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.73015046,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.7329912185668945
    },
    {
      "auxiliary_loss_clip": 0.01148814,
      "auxiliary_loss_mlp": 0.0102349,
      "balance_loss_clip": 1.04427004,
      "balance_loss_mlp": 1.01565766,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.7892941278747427,
      "language_loss": 0.69501007,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.7167331,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.6748855113983154
    },
    {
      "auxiliary_loss_clip": 0.01151895,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 1.04531288,
      "balance_loss_mlp": 1.01522315,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 3.066741224633953,
      "language_loss": 0.88896114,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.91070354,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.617170572280884
    },
    {
      "auxiliary_loss_clip": 0.01168621,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.05018067,
      "balance_loss_mlp": 1.01653349,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.796314550952503,
      "language_loss": 0.81731564,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83923519,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.5770461559295654
    },
    {
      "auxiliary_loss_clip": 0.01140639,
      "auxiliary_loss_mlp": 0.01025554,
      "balance_loss_clip": 1.04892409,
      "balance_loss_mlp": 1.01740646,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.6095485606443503,
      "language_loss": 0.75456953,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77623153,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.665499448776245
    },
    {
      "auxiliary_loss_clip": 0.01154539,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.05019093,
      "balance_loss_mlp": 1.01712203,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 3.500365683661326,
      "language_loss": 0.82054913,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84233713,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.638366460800171
    },
    {
      "auxiliary_loss_clip": 0.0116302,
      "auxiliary_loss_mlp": 0.01029265,
      "balance_loss_clip": 1.04639363,
      "balance_loss_mlp": 1.02202916,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 2.2649930086790935,
      "language_loss": 0.67543995,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69736278,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.6316487789154053
    },
    {
      "auxiliary_loss_clip": 0.01134239,
      "auxiliary_loss_mlp": 0.01022316,
      "balance_loss_clip": 1.03988266,
      "balance_loss_mlp": 1.01426911,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.9404862059178005,
      "language_loss": 0.65150619,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.6730718,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 3.6509571075439453
    },
    {
      "auxiliary_loss_clip": 0.01149977,
      "auxiliary_loss_mlp": 0.00900983,
      "balance_loss_clip": 1.0445044,
      "balance_loss_mlp": 1.00066948,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 3.534074917591796,
      "language_loss": 0.83603686,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.85654646,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.6686127185821533
    },
    {
      "auxiliary_loss_clip": 0.01161963,
      "auxiliary_loss_mlp": 0.01023027,
      "balance_loss_clip": 1.04582238,
      "balance_loss_mlp": 1.0161432,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 2.0495652226097025,
      "language_loss": 0.79650581,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81835568,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.6280667781829834
    },
    {
      "auxiliary_loss_clip": 0.01173644,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04895627,
      "balance_loss_mlp": 1.0195365,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 2.0229997245318003,
      "language_loss": 0.72813827,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75014722,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.6349501609802246
    },
    {
      "auxiliary_loss_clip": 0.01159041,
      "auxiliary_loss_mlp": 0.0102121,
      "balance_loss_clip": 1.05042291,
      "balance_loss_mlp": 1.0137887,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 2.3088563598130634,
      "language_loss": 0.88411188,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90591437,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.6576356887817383
    },
    {
      "auxiliary_loss_clip": 0.01141324,
      "auxiliary_loss_mlp": 0.01031368,
      "balance_loss_clip": 1.04168403,
      "balance_loss_mlp": 1.02248645,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.6600557587435527,
      "language_loss": 0.62607622,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.64780319,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.901589870452881
    },
    {
      "auxiliary_loss_clip": 0.01141134,
      "auxiliary_loss_mlp": 0.00901527,
      "balance_loss_clip": 1.04374313,
      "balance_loss_mlp": 1.00087583,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 2.0548494476406116,
      "language_loss": 0.75935692,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.77978349,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 3.6255202293395996
    },
    {
      "auxiliary_loss_clip": 0.01157345,
      "auxiliary_loss_mlp": 0.01022551,
      "balance_loss_clip": 1.04413414,
      "balance_loss_mlp": 1.01520765,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 2.257027140305466,
      "language_loss": 0.84230733,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.8641063,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.6361782550811768
    },
    {
      "auxiliary_loss_clip": 0.01137299,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.0442543,
      "balance_loss_mlp": 1.01687491,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 2.2513107559097834,
      "language_loss": 0.82355589,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84517223,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.7768783569335938
    },
    {
      "auxiliary_loss_clip": 0.0117209,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 1.0507195,
      "balance_loss_mlp": 1.02167606,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 1.9210529577153073,
      "language_loss": 0.70583814,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.72784805,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.5780065059661865
    },
    {
      "auxiliary_loss_clip": 0.01154042,
      "auxiliary_loss_mlp": 0.01025988,
      "balance_loss_clip": 1.046597,
      "balance_loss_mlp": 1.0182097,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.9606133466432298,
      "language_loss": 0.86573446,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.8875348,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 3.594067096710205
    },
    {
      "auxiliary_loss_clip": 0.01166745,
      "auxiliary_loss_mlp": 0.00901296,
      "balance_loss_clip": 1.04660273,
      "balance_loss_mlp": 1.00076532,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 2.147484728164326,
      "language_loss": 0.66018385,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.68086421,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.6661722660064697
    },
    {
      "auxiliary_loss_clip": 0.01147159,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.04639447,
      "balance_loss_mlp": 1.01812553,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 2.0133779109757293,
      "language_loss": 0.87667406,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89840257,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.642300605773926
    },
    {
      "auxiliary_loss_clip": 0.01156304,
      "auxiliary_loss_mlp": 0.01025865,
      "balance_loss_clip": 1.04972959,
      "balance_loss_mlp": 1.0184145,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 2.0763201274820484,
      "language_loss": 0.81054437,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.83236605,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.686265468597412
    },
    {
      "auxiliary_loss_clip": 0.01058928,
      "auxiliary_loss_mlp": 0.01002022,
      "balance_loss_clip": 1.01511061,
      "balance_loss_mlp": 1.00105083,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.81763134473312,
      "language_loss": 0.56427592,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58488542,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.278989553451538
    },
    {
      "auxiliary_loss_clip": 0.01157047,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.04842997,
      "balance_loss_mlp": 1.0168426,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 2.3911029549159792,
      "language_loss": 0.77567792,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79748666,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.663526773452759
    },
    {
      "auxiliary_loss_clip": 0.01149683,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.04601812,
      "balance_loss_mlp": 1.01866055,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 3.310926678483359,
      "language_loss": 0.67447889,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69623411,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.8080899715423584
    },
    {
      "auxiliary_loss_clip": 0.0114457,
      "auxiliary_loss_mlp": 0.0102399,
      "balance_loss_clip": 1.04088402,
      "balance_loss_mlp": 1.01659632,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.9340780881058182,
      "language_loss": 0.73428643,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75597203,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 3.7287309169769287
    },
    {
      "auxiliary_loss_clip": 0.01140646,
      "auxiliary_loss_mlp": 0.01026479,
      "balance_loss_clip": 1.04537868,
      "balance_loss_mlp": 1.01922488,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.9199492841288177,
      "language_loss": 0.85002464,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87169588,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.7087764739990234
    },
    {
      "auxiliary_loss_clip": 0.01169114,
      "auxiliary_loss_mlp": 0.01023807,
      "balance_loss_clip": 1.05069947,
      "balance_loss_mlp": 1.01682436,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 2.5785769356415598,
      "language_loss": 0.74066162,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.76259089,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.615697145462036
    },
    {
      "auxiliary_loss_clip": 0.01151368,
      "auxiliary_loss_mlp": 0.01025717,
      "balance_loss_clip": 1.05015707,
      "balance_loss_mlp": 1.01696384,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 6.364406957928854,
      "language_loss": 0.73636067,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.7581315,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.647986650466919
    },
    {
      "auxiliary_loss_clip": 0.01125383,
      "auxiliary_loss_mlp": 0.01024988,
      "balance_loss_clip": 1.03841305,
      "balance_loss_mlp": 1.01778793,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.9029996666149083,
      "language_loss": 0.68974018,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71124387,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.750840187072754
    },
    {
      "auxiliary_loss_clip": 0.01139585,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04836273,
      "balance_loss_mlp": 1.02161109,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 2.2820091368402866,
      "language_loss": 0.7768476,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.79853678,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.852400302886963
    },
    {
      "auxiliary_loss_clip": 0.01134479,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.0382247,
      "balance_loss_mlp": 1.01927221,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 2.065552719198726,
      "language_loss": 0.69835973,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71996868,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.831820011138916
    },
    {
      "auxiliary_loss_clip": 0.01071506,
      "auxiliary_loss_mlp": 0.01000983,
      "balance_loss_clip": 1.01202238,
      "balance_loss_mlp": 1.0001421,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8518410287280587,
      "language_loss": 0.66254842,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68327332,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.2924563884735107
    },
    {
      "auxiliary_loss_clip": 0.01159125,
      "auxiliary_loss_mlp": 0.00901221,
      "balance_loss_clip": 1.04727507,
      "balance_loss_mlp": 1.00081134,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.86931616025297,
      "language_loss": 0.64991415,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.67051768,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.6505630016326904
    },
    {
      "auxiliary_loss_clip": 0.01159291,
      "auxiliary_loss_mlp": 0.01023436,
      "balance_loss_clip": 1.04706931,
      "balance_loss_mlp": 1.01611686,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.0212639077548826,
      "language_loss": 0.87079102,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89261824,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 3.63468074798584
    },
    {
      "auxiliary_loss_clip": 0.01071488,
      "auxiliary_loss_mlp": 0.01001146,
      "balance_loss_clip": 1.01160669,
      "balance_loss_mlp": 1.00034177,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.841179966645142,
      "language_loss": 0.60345566,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62418199,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.11092209815979
    },
    {
      "auxiliary_loss_clip": 0.01149885,
      "auxiliary_loss_mlp": 0.01027654,
      "balance_loss_clip": 1.0439353,
      "balance_loss_mlp": 1.02066255,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 2.3656025803809073,
      "language_loss": 0.76102471,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78280014,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.7451019287109375
    },
    {
      "auxiliary_loss_clip": 0.01150702,
      "auxiliary_loss_mlp": 0.01022444,
      "balance_loss_clip": 1.04441404,
      "balance_loss_mlp": 1.01484776,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.333022186670456,
      "language_loss": 0.70512599,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72685754,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.61623215675354
    },
    {
      "auxiliary_loss_clip": 0.01139294,
      "auxiliary_loss_mlp": 0.01024637,
      "balance_loss_clip": 1.04405808,
      "balance_loss_mlp": 1.01708233,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 1.8667276760965035,
      "language_loss": 0.83875179,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86039108,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.709535837173462
    },
    {
      "auxiliary_loss_clip": 0.01144461,
      "auxiliary_loss_mlp": 0.01026733,
      "balance_loss_clip": 1.04419017,
      "balance_loss_mlp": 1.01845384,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.3981366677119325,
      "language_loss": 0.73787928,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75959122,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.7430896759033203
    },
    {
      "auxiliary_loss_clip": 0.01151526,
      "auxiliary_loss_mlp": 0.01023692,
      "balance_loss_clip": 1.0477289,
      "balance_loss_mlp": 1.01588392,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.600084450881969,
      "language_loss": 0.78050375,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.80225599,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 3.594413995742798
    },
    {
      "auxiliary_loss_clip": 0.01173855,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.0495677,
      "balance_loss_mlp": 1.01706171,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.7328953256965938,
      "language_loss": 0.76564604,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.78763223,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.580754041671753
    },
    {
      "auxiliary_loss_clip": 0.0114563,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.04365194,
      "balance_loss_mlp": 1.02113628,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.4126906656331424,
      "language_loss": 0.71373886,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73548079,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.7024266719818115
    },
    {
      "auxiliary_loss_clip": 0.01159905,
      "auxiliary_loss_mlp": 0.01020929,
      "balance_loss_clip": 1.04803228,
      "balance_loss_mlp": 1.01348448,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.7064340355794763,
      "language_loss": 0.76496178,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78677011,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.5978736877441406
    },
    {
      "auxiliary_loss_clip": 0.01130564,
      "auxiliary_loss_mlp": 0.01025475,
      "balance_loss_clip": 1.04210639,
      "balance_loss_mlp": 1.01766717,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 2.299970258545814,
      "language_loss": 0.64545602,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66701639,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 3.6545021533966064
    },
    {
      "auxiliary_loss_clip": 0.01171817,
      "auxiliary_loss_mlp": 0.01020348,
      "balance_loss_clip": 1.05035877,
      "balance_loss_mlp": 1.01307046,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.6659111719611657,
      "language_loss": 0.86360723,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88552886,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.5759503841400146
    },
    {
      "auxiliary_loss_clip": 0.01139013,
      "auxiliary_loss_mlp": 0.01022488,
      "balance_loss_clip": 1.04678512,
      "balance_loss_mlp": 1.01556778,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 2.3899021006632926,
      "language_loss": 0.69638771,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.7180028,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.7239696979522705
    },
    {
      "auxiliary_loss_clip": 0.01163601,
      "auxiliary_loss_mlp": 0.01025362,
      "balance_loss_clip": 1.04980278,
      "balance_loss_mlp": 1.01834106,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 1.8246032909398873,
      "language_loss": 0.82125014,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84313977,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.6472582817077637
    },
    {
      "auxiliary_loss_clip": 0.01059571,
      "auxiliary_loss_mlp": 0.00890846,
      "balance_loss_clip": 1.01155603,
      "balance_loss_mlp": 0.99986011,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6579906901436801,
      "language_loss": 0.56233418,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58183831,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.196864128112793
    },
    {
      "auxiliary_loss_clip": 0.01154141,
      "auxiliary_loss_mlp": 0.01020072,
      "balance_loss_clip": 1.04487455,
      "balance_loss_mlp": 1.01278853,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 2.130608638389408,
      "language_loss": 0.67339599,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69513816,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.6317734718322754
    },
    {
      "auxiliary_loss_clip": 0.01149395,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.04653549,
      "balance_loss_mlp": 1.02231491,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.3216865879328275,
      "language_loss": 0.713947,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73574173,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.6488771438598633
    },
    {
      "auxiliary_loss_clip": 0.0115575,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.04602432,
      "balance_loss_mlp": 1.01976132,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.6120067186158935,
      "language_loss": 0.75394022,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77577364,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.6741795539855957
    },
    {
      "auxiliary_loss_clip": 0.01139859,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.04291511,
      "balance_loss_mlp": 1.02120328,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 8.762786476336931,
      "language_loss": 0.80647337,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82816273,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 3.692701578140259
    },
    {
      "auxiliary_loss_clip": 0.01170627,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.05042803,
      "balance_loss_mlp": 1.01899672,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 1.8626393367191432,
      "language_loss": 0.80477649,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82674336,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.594560146331787
    },
    {
      "auxiliary_loss_clip": 0.01164567,
      "auxiliary_loss_mlp": 0.01033246,
      "balance_loss_clip": 1.04989672,
      "balance_loss_mlp": 1.02587306,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 2.1437098515261486,
      "language_loss": 0.7514531,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.7734313,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.6638054847717285
    },
    {
      "auxiliary_loss_clip": 0.01175604,
      "auxiliary_loss_mlp": 0.01031062,
      "balance_loss_clip": 1.05019283,
      "balance_loss_mlp": 1.02258003,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.9059728422443716,
      "language_loss": 0.66858792,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.69065458,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.5360167026519775
    },
    {
      "auxiliary_loss_clip": 0.01166439,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.04690158,
      "balance_loss_mlp": 1.02052414,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.685432494479866,
      "language_loss": 0.78519726,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80713761,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.6871249675750732
    },
    {
      "auxiliary_loss_clip": 0.01156957,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.04487908,
      "balance_loss_mlp": 1.02037632,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 2.2123994716514837,
      "language_loss": 0.71433711,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73618257,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.8413076400756836
    },
    {
      "auxiliary_loss_clip": 0.01045061,
      "auxiliary_loss_mlp": 0.01002016,
      "balance_loss_clip": 1.00953984,
      "balance_loss_mlp": 1.00107992,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7340861844931102,
      "language_loss": 0.55411303,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57458377,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.118424415588379
    },
    {
      "auxiliary_loss_clip": 0.01044011,
      "auxiliary_loss_mlp": 0.0100202,
      "balance_loss_clip": 1.01666439,
      "balance_loss_mlp": 1.00113797,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7800201165576287,
      "language_loss": 0.64692825,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66738856,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.18027925491333
    },
    {
      "auxiliary_loss_clip": 0.01171185,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.05017114,
      "balance_loss_mlp": 1.0188458,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 2.266843336864232,
      "language_loss": 0.7295481,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.75151742,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.5095434188842773
    },
    {
      "auxiliary_loss_clip": 0.01158286,
      "auxiliary_loss_mlp": 0.01025365,
      "balance_loss_clip": 1.04716086,
      "balance_loss_mlp": 1.01829624,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 2.1647021095873056,
      "language_loss": 0.81623447,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83807093,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 3.626770496368408
    },
    {
      "auxiliary_loss_clip": 0.01152387,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.04679501,
      "balance_loss_mlp": 1.01768088,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.9764687582928684,
      "language_loss": 0.72137785,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74315417,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 2.6573452949523926
    },
    {
      "auxiliary_loss_clip": 0.01127149,
      "auxiliary_loss_mlp": 0.01028467,
      "balance_loss_clip": 1.04364395,
      "balance_loss_mlp": 1.01985991,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 1.9925538785079877,
      "language_loss": 0.78148341,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80303955,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.7469048500061035
    },
    {
      "auxiliary_loss_clip": 0.01171166,
      "auxiliary_loss_mlp": 0.01026484,
      "balance_loss_clip": 1.04837215,
      "balance_loss_mlp": 1.01843739,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 2.220925933559928,
      "language_loss": 0.74618137,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.76815784,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.535336971282959
    },
    {
      "auxiliary_loss_clip": 0.01051687,
      "auxiliary_loss_mlp": 0.00890973,
      "balance_loss_clip": 1.01123929,
      "balance_loss_mlp": 1.00000751,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9498140620807138,
      "language_loss": 0.66203225,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68145883,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.3137295246124268
    },
    {
      "auxiliary_loss_clip": 0.01170761,
      "auxiliary_loss_mlp": 0.01025901,
      "balance_loss_clip": 1.04912388,
      "balance_loss_mlp": 1.01871872,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.8054942168977943,
      "language_loss": 0.73098695,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75295353,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 3.5926082134246826
    },
    {
      "auxiliary_loss_clip": 0.01143156,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.04705966,
      "balance_loss_mlp": 1.01938653,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.769327390509695,
      "language_loss": 0.75331903,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77501845,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.699497699737549
    },
    {
      "auxiliary_loss_clip": 0.01171897,
      "auxiliary_loss_mlp": 0.01028689,
      "balance_loss_clip": 1.05032384,
      "balance_loss_mlp": 1.02072537,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.752761516135375,
      "language_loss": 0.58365333,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60565925,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.5685503482818604
    },
    {
      "auxiliary_loss_clip": 0.01173044,
      "auxiliary_loss_mlp": 0.00901403,
      "balance_loss_clip": 1.05033028,
      "balance_loss_mlp": 1.00088954,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 1.9149367807711026,
      "language_loss": 0.71970749,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.74045193,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.5342273712158203
    },
    {
      "auxiliary_loss_clip": 0.0117179,
      "auxiliary_loss_mlp": 0.01020291,
      "balance_loss_clip": 1.04915452,
      "balance_loss_mlp": 1.01319492,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 3.2733615853713047,
      "language_loss": 0.53829122,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.56021202,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.553173303604126
    },
    {
      "auxiliary_loss_clip": 0.01162449,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.04840505,
      "balance_loss_mlp": 1.02305043,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.6020033738947979,
      "language_loss": 0.65720952,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67913973,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 3.8011159896850586
    },
    {
      "auxiliary_loss_clip": 0.01159762,
      "auxiliary_loss_mlp": 0.01026493,
      "balance_loss_clip": 1.04500806,
      "balance_loss_mlp": 1.01923966,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.799012135209475,
      "language_loss": 0.75612688,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.77798939,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.54695725440979
    },
    {
      "auxiliary_loss_clip": 0.0115932,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 1.04732645,
      "balance_loss_mlp": 1.02074707,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 4.198351514080207,
      "language_loss": 0.83263546,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85450661,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.6048405170440674
    },
    {
      "auxiliary_loss_clip": 0.01070103,
      "auxiliary_loss_mlp": 0.01000749,
      "balance_loss_clip": 1.013659,
      "balance_loss_mlp": 0.99991494,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7866471018907994,
      "language_loss": 0.56748068,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58818918,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.126661777496338
    },
    {
      "auxiliary_loss_clip": 0.01153918,
      "auxiliary_loss_mlp": 0.01023695,
      "balance_loss_clip": 1.04743159,
      "balance_loss_mlp": 1.01615834,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 3.181561766114851,
      "language_loss": 0.63365936,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.6554355,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.601823091506958
    },
    {
      "auxiliary_loss_clip": 0.01147545,
      "auxiliary_loss_mlp": 0.01033669,
      "balance_loss_clip": 1.04187763,
      "balance_loss_mlp": 1.02618015,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.221317214555746,
      "language_loss": 0.77143753,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79324967,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.6770262718200684
    },
    {
      "auxiliary_loss_clip": 0.0116598,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.04793811,
      "balance_loss_mlp": 1.01656008,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.6612558483907505,
      "language_loss": 0.7996766,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82157302,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.6036951541900635
    },
    {
      "auxiliary_loss_clip": 0.01067178,
      "auxiliary_loss_mlp": 0.01001503,
      "balance_loss_clip": 1.01158404,
      "balance_loss_mlp": 1.00065029,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8103459748936083,
      "language_loss": 0.63915092,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.65983778,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.0467135906219482
    },
    {
      "auxiliary_loss_clip": 0.01163241,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 1.04955065,
      "balance_loss_mlp": 1.01938617,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.2541631360207917,
      "language_loss": 0.74161935,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.76351255,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 3.533726930618286
    },
    {
      "auxiliary_loss_clip": 0.01168853,
      "auxiliary_loss_mlp": 0.01022607,
      "balance_loss_clip": 1.04822242,
      "balance_loss_mlp": 1.01566958,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.0110065041028955,
      "language_loss": 0.76819062,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.79010528,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.541867733001709
    },
    {
      "auxiliary_loss_clip": 0.01038407,
      "auxiliary_loss_mlp": 0.01002185,
      "balance_loss_clip": 1.01029563,
      "balance_loss_mlp": 1.0012846,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8034306720002424,
      "language_loss": 0.62381798,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64422387,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.424898386001587
    },
    {
      "auxiliary_loss_clip": 0.0106498,
      "auxiliary_loss_mlp": 0.01001225,
      "balance_loss_clip": 1.0119524,
      "balance_loss_mlp": 1.00041413,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9346963456790711,
      "language_loss": 0.57788593,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.598548,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.293856382369995
    },
    {
      "auxiliary_loss_clip": 0.01169893,
      "auxiliary_loss_mlp": 0.01024683,
      "balance_loss_clip": 1.04742265,
      "balance_loss_mlp": 1.01738167,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.910096918596133,
      "language_loss": 0.76559997,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78754568,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.74261736869812
    },
    {
      "auxiliary_loss_clip": 0.01177057,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 1.0509088,
      "balance_loss_mlp": 1.02474177,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.7865262892377296,
      "language_loss": 0.72895885,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75105023,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.8570942878723145
    },
    {
      "auxiliary_loss_clip": 0.01161276,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.05081534,
      "balance_loss_mlp": 1.01904058,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.031627638630271,
      "language_loss": 0.66580129,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68767846,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.6967902183532715
    },
    {
      "auxiliary_loss_clip": 0.01137021,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 1.04016566,
      "balance_loss_mlp": 1.01768148,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.6026407667162252,
      "language_loss": 0.72342062,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.7450453,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.7361698150634766
    },
    {
      "auxiliary_loss_clip": 0.01162531,
      "auxiliary_loss_mlp": 0.01024832,
      "balance_loss_clip": 1.04956913,
      "balance_loss_mlp": 1.01792121,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 4.358828427576338,
      "language_loss": 0.79591829,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81779188,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.66947603225708
    },
    {
      "auxiliary_loss_clip": 0.01138865,
      "auxiliary_loss_mlp": 0.01020256,
      "balance_loss_clip": 1.04331923,
      "balance_loss_mlp": 1.01322842,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.7962773562583565,
      "language_loss": 0.74240732,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76399851,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 3.6913230419158936
    },
    {
      "auxiliary_loss_clip": 0.01141906,
      "auxiliary_loss_mlp": 0.01029026,
      "balance_loss_clip": 1.043468,
      "balance_loss_mlp": 1.02159023,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 1.7572478433287466,
      "language_loss": 0.855407,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87711632,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 2.624032974243164
    },
    {
      "auxiliary_loss_clip": 0.01155227,
      "auxiliary_loss_mlp": 0.01023899,
      "balance_loss_clip": 1.04816842,
      "balance_loss_mlp": 1.01742589,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.742226670819209,
      "language_loss": 0.81168532,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.8334766,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.6953704357147217
    },
    {
      "auxiliary_loss_clip": 0.01156999,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.04907084,
      "balance_loss_mlp": 1.02193069,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 1.8173665231711276,
      "language_loss": 0.74926496,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.77113473,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.6068553924560547
    },
    {
      "auxiliary_loss_clip": 0.01056759,
      "auxiliary_loss_mlp": 0.01004316,
      "balance_loss_clip": 1.01294172,
      "balance_loss_mlp": 1.00334477,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7519774680679285,
      "language_loss": 0.58974707,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61035788,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.2966105937957764
    },
    {
      "auxiliary_loss_clip": 0.01159759,
      "auxiliary_loss_mlp": 0.01026277,
      "balance_loss_clip": 1.04637766,
      "balance_loss_mlp": 1.01876092,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.225112079179095,
      "language_loss": 0.7883755,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.8102358,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.5742123126983643
    },
    {
      "auxiliary_loss_clip": 0.01165202,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.04753757,
      "balance_loss_mlp": 1.02075303,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.8647290466551187,
      "language_loss": 0.76391661,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78585124,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 3.5621156692504883
    },
    {
      "auxiliary_loss_clip": 0.01129516,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04541278,
      "balance_loss_mlp": 1.01867867,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 1.8472392443072307,
      "language_loss": 0.64575875,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66731912,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.8173956871032715
    },
    {
      "auxiliary_loss_clip": 0.01164683,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04946291,
      "balance_loss_mlp": 1.01820111,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 2.208342716825926,
      "language_loss": 0.83156544,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85346889,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.619910717010498
    },
    {
      "auxiliary_loss_clip": 0.01173666,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.04909074,
      "balance_loss_mlp": 1.02125132,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 2.038850822469059,
      "language_loss": 0.77266514,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79469264,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.6330699920654297
    },
    {
      "auxiliary_loss_clip": 0.01175293,
      "auxiliary_loss_mlp": 0.01023792,
      "balance_loss_clip": 1.04950893,
      "balance_loss_mlp": 1.01623762,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 1.6728063022748128,
      "language_loss": 0.67238104,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69437194,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 3.5404672622680664
    },
    {
      "auxiliary_loss_clip": 0.0114856,
      "auxiliary_loss_mlp": 0.01024947,
      "balance_loss_clip": 1.04438746,
      "balance_loss_mlp": 1.01752663,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.2599088622188614,
      "language_loss": 0.66238457,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68411958,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.7841763496398926
    },
    {
      "auxiliary_loss_clip": 0.0117009,
      "auxiliary_loss_mlp": 0.01021619,
      "balance_loss_clip": 1.04664838,
      "balance_loss_mlp": 1.01373327,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.8272428826289544,
      "language_loss": 0.79613435,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81805146,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.5394883155822754
    },
    {
      "auxiliary_loss_clip": 0.0116131,
      "auxiliary_loss_mlp": 0.01023073,
      "balance_loss_clip": 1.04699337,
      "balance_loss_mlp": 1.01582813,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.48255032839378,
      "language_loss": 0.73244458,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75428843,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.566134214401245
    },
    {
      "auxiliary_loss_clip": 0.01128056,
      "auxiliary_loss_mlp": 0.01022717,
      "balance_loss_clip": 1.03856945,
      "balance_loss_mlp": 1.01549315,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.502270940170047,
      "language_loss": 0.76843762,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.78994536,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.676758050918579
    },
    {
      "auxiliary_loss_clip": 0.01157953,
      "auxiliary_loss_mlp": 0.01020378,
      "balance_loss_clip": 1.04702091,
      "balance_loss_mlp": 1.01366603,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 2.242324480314996,
      "language_loss": 0.90329975,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92508304,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.743781328201294
    },
    {
      "auxiliary_loss_clip": 0.01168587,
      "auxiliary_loss_mlp": 0.01023578,
      "balance_loss_clip": 1.04723382,
      "balance_loss_mlp": 1.01659799,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.7845985836823846,
      "language_loss": 0.80304527,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82496691,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.606701612472534
    },
    {
      "auxiliary_loss_clip": 0.01159793,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.0466001,
      "balance_loss_mlp": 1.01702964,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 3.709840708053364,
      "language_loss": 0.78223747,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80407971,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 3.536867141723633
    },
    {
      "auxiliary_loss_clip": 0.01157933,
      "auxiliary_loss_mlp": 0.00901684,
      "balance_loss_clip": 1.04464257,
      "balance_loss_mlp": 1.0008285,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 1.788460671908692,
      "language_loss": 0.66514075,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68573689,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.6201274394989014
    },
    {
      "auxiliary_loss_clip": 0.01158484,
      "auxiliary_loss_mlp": 0.01026185,
      "balance_loss_clip": 1.04483581,
      "balance_loss_mlp": 1.01872838,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.7284488057797973,
      "language_loss": 0.88761246,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.90945911,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.6631109714508057
    },
    {
      "auxiliary_loss_clip": 0.01159307,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.04854274,
      "balance_loss_mlp": 1.01841009,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.680080731303603,
      "language_loss": 0.79105866,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81291354,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.6326382160186768
    },
    {
      "auxiliary_loss_clip": 0.01164817,
      "auxiliary_loss_mlp": 0.009016,
      "balance_loss_clip": 1.05422163,
      "balance_loss_mlp": 1.00083041,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.736833138053528,
      "language_loss": 0.8578825,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.87854671,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.6306264400482178
    },
    {
      "auxiliary_loss_clip": 0.01151681,
      "auxiliary_loss_mlp": 0.01028204,
      "balance_loss_clip": 1.04592705,
      "balance_loss_mlp": 1.02070248,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 1.773725692411665,
      "language_loss": 0.87353289,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89533174,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.599287509918213
    },
    {
      "auxiliary_loss_clip": 0.0114842,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.04359007,
      "balance_loss_mlp": 1.02328956,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.8089432232686686,
      "language_loss": 0.77593517,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79772782,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.6022696495056152
    },
    {
      "auxiliary_loss_clip": 0.01154306,
      "auxiliary_loss_mlp": 0.01029677,
      "balance_loss_clip": 1.04508317,
      "balance_loss_mlp": 1.02217865,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.955162874611141,
      "language_loss": 0.85197008,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87380993,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.655881404876709
    },
    {
      "auxiliary_loss_clip": 0.01174424,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05110562,
      "balance_loss_mlp": 1.02224874,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.3090735230036805,
      "language_loss": 0.86803973,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.89007574,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.551760673522949
    },
    {
      "auxiliary_loss_clip": 0.01154303,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 1.04496074,
      "balance_loss_mlp": 1.02050757,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.0470696590507638,
      "language_loss": 0.66799355,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68981725,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 2.687274932861328
    },
    {
      "auxiliary_loss_clip": 0.01145798,
      "auxiliary_loss_mlp": 0.01028663,
      "balance_loss_clip": 1.04318595,
      "balance_loss_mlp": 1.02068245,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.682695328638998,
      "language_loss": 0.6883781,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.7101227,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 3.578618288040161
    },
    {
      "auxiliary_loss_clip": 0.01147511,
      "auxiliary_loss_mlp": 0.01030526,
      "balance_loss_clip": 1.04619122,
      "balance_loss_mlp": 1.02275324,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 2.1849698833129225,
      "language_loss": 0.79362917,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81540954,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.747544288635254
    },
    {
      "auxiliary_loss_clip": 0.01150801,
      "auxiliary_loss_mlp": 0.01022582,
      "balance_loss_clip": 1.04809332,
      "balance_loss_mlp": 1.01578724,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 2.270449468111797,
      "language_loss": 0.84684706,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.86858094,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.6864709854125977
    },
    {
      "auxiliary_loss_clip": 0.01150342,
      "auxiliary_loss_mlp": 0.01024996,
      "balance_loss_clip": 1.04517925,
      "balance_loss_mlp": 1.01739907,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 2.340251914576459,
      "language_loss": 0.75234067,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77409405,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.590965747833252
    },
    {
      "auxiliary_loss_clip": 0.01135434,
      "auxiliary_loss_mlp": 0.01029522,
      "balance_loss_clip": 1.04113245,
      "balance_loss_mlp": 1.02166033,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.7965108055796255,
      "language_loss": 0.72167593,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74332547,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.7404658794403076
    },
    {
      "auxiliary_loss_clip": 0.01045871,
      "auxiliary_loss_mlp": 0.01000789,
      "balance_loss_clip": 1.01260352,
      "balance_loss_mlp": 1.0000205,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.7086183088265688,
      "language_loss": 0.56563139,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58609796,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 4.448968410491943
    },
    {
      "auxiliary_loss_clip": 0.01066471,
      "auxiliary_loss_mlp": 0.01000895,
      "balance_loss_clip": 1.01141942,
      "balance_loss_mlp": 1.00004828,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.774277429335025,
      "language_loss": 0.54686975,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56754339,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.3374125957489014
    },
    {
      "auxiliary_loss_clip": 0.01128542,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.04391289,
      "balance_loss_mlp": 1.01879466,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.142856582352424,
      "language_loss": 0.79191339,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81346333,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.706275463104248
    },
    {
      "auxiliary_loss_clip": 0.01165164,
      "auxiliary_loss_mlp": 0.00901165,
      "balance_loss_clip": 1.04815388,
      "balance_loss_mlp": 1.00078571,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 1.8915629375778644,
      "language_loss": 0.74805266,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76871598,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.651897668838501
    },
    {
      "auxiliary_loss_clip": 0.01143139,
      "auxiliary_loss_mlp": 0.01022053,
      "balance_loss_clip": 1.0474267,
      "balance_loss_mlp": 1.01479328,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 2.0205944322122606,
      "language_loss": 0.83772063,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.8593725,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.6432595252990723
    },
    {
      "auxiliary_loss_clip": 0.01165124,
      "auxiliary_loss_mlp": 0.00901413,
      "balance_loss_clip": 1.04680145,
      "balance_loss_mlp": 1.00084591,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 2.0516400230210716,
      "language_loss": 0.84434128,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.86500669,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 3.6119372844696045
    },
    {
      "auxiliary_loss_clip": 0.01075936,
      "auxiliary_loss_mlp": 0.01003395,
      "balance_loss_clip": 1.02861595,
      "balance_loss_mlp": 1.00248289,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.796989941947562,
      "language_loss": 0.56440794,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58520126,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.1952621936798096
    },
    {
      "auxiliary_loss_clip": 0.0114846,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04492211,
      "balance_loss_mlp": 1.01711547,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.5693773212921367,
      "language_loss": 0.78024262,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.8019712,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.685502052307129
    },
    {
      "auxiliary_loss_clip": 0.01147002,
      "auxiliary_loss_mlp": 0.01027364,
      "balance_loss_clip": 1.04559505,
      "balance_loss_mlp": 1.02007174,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.036693192965846,
      "language_loss": 0.7271843,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74892789,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.5988428592681885
    },
    {
      "auxiliary_loss_clip": 0.01157698,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.0456897,
      "balance_loss_mlp": 1.02059186,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 2.1635487198056262,
      "language_loss": 0.67973411,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70159054,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.5760395526885986
    },
    {
      "auxiliary_loss_clip": 0.0116935,
      "auxiliary_loss_mlp": 0.01023489,
      "balance_loss_clip": 1.04650116,
      "balance_loss_mlp": 1.01647377,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.540328657464548,
      "language_loss": 0.70942938,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.73135781,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.6938893795013428
    },
    {
      "auxiliary_loss_clip": 0.01141038,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.04438198,
      "balance_loss_mlp": 1.01933765,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.8931160820703392,
      "language_loss": 0.74376428,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76544112,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 3.6487877368927
    },
    {
      "auxiliary_loss_clip": 0.01061296,
      "auxiliary_loss_mlp": 0.01004371,
      "balance_loss_clip": 1.01188874,
      "balance_loss_mlp": 1.00344741,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7810889551827584,
      "language_loss": 0.52936739,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.55002409,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.1237082481384277
    },
    {
      "auxiliary_loss_clip": 0.01133946,
      "auxiliary_loss_mlp": 0.0102379,
      "balance_loss_clip": 1.04297769,
      "balance_loss_mlp": 1.01587474,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 2.358030147289335,
      "language_loss": 0.71052253,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73209977,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.667980909347534
    },
    {
      "auxiliary_loss_clip": 0.01153731,
      "auxiliary_loss_mlp": 0.01028001,
      "balance_loss_clip": 1.04699874,
      "balance_loss_mlp": 1.02017486,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.180406515630372,
      "language_loss": 0.73180789,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75362521,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.6552202701568604
    },
    {
      "auxiliary_loss_clip": 0.0115264,
      "auxiliary_loss_mlp": 0.01021247,
      "balance_loss_clip": 1.04157448,
      "balance_loss_mlp": 1.01412416,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 2.187431149397185,
      "language_loss": 0.7394346,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76117349,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.6235601902008057
    },
    {
      "auxiliary_loss_clip": 0.01155274,
      "auxiliary_loss_mlp": 0.01024129,
      "balance_loss_clip": 1.04566038,
      "balance_loss_mlp": 1.01708341,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.4376771145180742,
      "language_loss": 0.79412222,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81591618,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.6339306831359863
    },
    {
      "auxiliary_loss_clip": 0.01170098,
      "auxiliary_loss_mlp": 0.00901517,
      "balance_loss_clip": 1.04869998,
      "balance_loss_mlp": 1.00080538,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.82021681502965,
      "language_loss": 0.66373301,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68444914,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.6036806106567383
    },
    {
      "auxiliary_loss_clip": 0.01153733,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.04527938,
      "balance_loss_mlp": 1.01841855,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.3180617852459475,
      "language_loss": 0.78258848,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80438733,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.6431431770324707
    },
    {
      "auxiliary_loss_clip": 0.01158981,
      "auxiliary_loss_mlp": 0.01025784,
      "balance_loss_clip": 1.04816246,
      "balance_loss_mlp": 1.01852155,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.3847639783481958,
      "language_loss": 0.7652235,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78707117,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.620440721511841
    },
    {
      "auxiliary_loss_clip": 0.01161027,
      "auxiliary_loss_mlp": 0.0102471,
      "balance_loss_clip": 1.04773092,
      "balance_loss_mlp": 1.01751614,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 1.7806980724763684,
      "language_loss": 0.70337564,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72523302,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 2.8924243450164795
    },
    {
      "auxiliary_loss_clip": 0.01154359,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.04368293,
      "balance_loss_mlp": 1.01872754,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.8112233557337187,
      "language_loss": 0.76488566,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78669715,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 3.910635232925415
    },
    {
      "auxiliary_loss_clip": 0.01171741,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.04867101,
      "balance_loss_mlp": 1.01750219,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.9581581881812715,
      "language_loss": 0.80539918,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82736689,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.614805221557617
    },
    {
      "auxiliary_loss_clip": 0.01157247,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04919755,
      "balance_loss_mlp": 1.01665592,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.439101453585754,
      "language_loss": 0.61760187,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.6394186,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.6925926208496094
    },
    {
      "auxiliary_loss_clip": 0.01143627,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04256165,
      "balance_loss_mlp": 1.01824117,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.7057231115525906,
      "language_loss": 0.74333811,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76503205,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.7781543731689453
    },
    {
      "auxiliary_loss_clip": 0.01169598,
      "auxiliary_loss_mlp": 0.01022557,
      "balance_loss_clip": 1.04670119,
      "balance_loss_mlp": 1.01553524,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 7.875974905888651,
      "language_loss": 0.79118955,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81311107,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.6239805221557617
    },
    {
      "auxiliary_loss_clip": 0.01170633,
      "auxiliary_loss_mlp": 0.01023189,
      "balance_loss_clip": 1.04789913,
      "balance_loss_mlp": 1.01595616,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 1.603564319227211,
      "language_loss": 0.79661095,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81854916,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.5655040740966797
    },
    {
      "auxiliary_loss_clip": 0.01157436,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.04852271,
      "balance_loss_mlp": 1.02573204,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.6123313688957355,
      "language_loss": 0.72934604,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.75126219,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 3.5820257663726807
    },
    {
      "auxiliary_loss_clip": 0.01165834,
      "auxiliary_loss_mlp": 0.01031274,
      "balance_loss_clip": 1.04824305,
      "balance_loss_mlp": 1.02351379,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 2.040738979992493,
      "language_loss": 0.78395832,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80592936,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.6460137367248535
    },
    {
      "auxiliary_loss_clip": 0.01162851,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.04948795,
      "balance_loss_mlp": 1.0197351,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.0973744659662947,
      "language_loss": 0.71121275,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.73311412,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.658034563064575
    },
    {
      "auxiliary_loss_clip": 0.01147923,
      "auxiliary_loss_mlp": 0.01021099,
      "balance_loss_clip": 1.04501414,
      "balance_loss_mlp": 1.01292133,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 1.630047354886883,
      "language_loss": 0.65653336,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.67822361,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.705129384994507
    },
    {
      "auxiliary_loss_clip": 0.01158988,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.0468117,
      "balance_loss_mlp": 1.0179522,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 2.7136793522357445,
      "language_loss": 0.76896071,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.79080331,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 3.573812484741211
    },
    {
      "auxiliary_loss_clip": 0.0115217,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.04438233,
      "balance_loss_mlp": 1.02041745,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.8620846375136515,
      "language_loss": 0.7222684,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74407095,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.642188310623169
    },
    {
      "auxiliary_loss_clip": 0.0112054,
      "auxiliary_loss_mlp": 0.01025944,
      "balance_loss_clip": 1.04442,
      "balance_loss_mlp": 1.0179925,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.9640017332616004,
      "language_loss": 0.81662083,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.83808571,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.8544986248016357
    },
    {
      "auxiliary_loss_clip": 0.01154265,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.04890847,
      "balance_loss_mlp": 1.01732302,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.9842591492619772,
      "language_loss": 0.82117814,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.8429687,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.615018844604492
    },
    {
      "auxiliary_loss_clip": 0.01148719,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.04874969,
      "balance_loss_mlp": 1.01999378,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.8755506035837257,
      "language_loss": 0.79142034,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.81317896,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.7077600955963135
    },
    {
      "auxiliary_loss_clip": 0.01173061,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 1.04774308,
      "balance_loss_mlp": 1.01525533,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.074177414971008,
      "language_loss": 0.7937417,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81570095,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.537498950958252
    },
    {
      "auxiliary_loss_clip": 0.01137496,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.0416975,
      "balance_loss_mlp": 1.0176605,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.730106630040688,
      "language_loss": 0.69762313,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.71924806,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.8676509857177734
    },
    {
      "auxiliary_loss_clip": 0.01154456,
      "auxiliary_loss_mlp": 0.01022998,
      "balance_loss_clip": 1.04918909,
      "balance_loss_mlp": 1.01622415,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.7735486347097382,
      "language_loss": 0.76171935,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78349388,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 3.669020175933838
    },
    {
      "auxiliary_loss_clip": 0.01157574,
      "auxiliary_loss_mlp": 0.0102133,
      "balance_loss_clip": 1.04550016,
      "balance_loss_mlp": 1.01435328,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.8815099072439392,
      "language_loss": 0.78299183,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.8047809,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.64210844039917
    },
    {
      "auxiliary_loss_clip": 0.01148988,
      "auxiliary_loss_mlp": 0.01024214,
      "balance_loss_clip": 1.04650629,
      "balance_loss_mlp": 1.01689792,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 2.176675705524086,
      "language_loss": 0.72120595,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74293798,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.6530120372772217
    },
    {
      "auxiliary_loss_clip": 0.01156267,
      "auxiliary_loss_mlp": 0.01022927,
      "balance_loss_clip": 1.04458809,
      "balance_loss_mlp": 1.01542878,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.8116078094339707,
      "language_loss": 0.74395895,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76575089,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.8643319606781006
    },
    {
      "auxiliary_loss_clip": 0.0112576,
      "auxiliary_loss_mlp": 0.01022864,
      "balance_loss_clip": 1.03768754,
      "balance_loss_mlp": 1.01518071,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.5171474431550251,
      "language_loss": 0.85868931,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.88017547,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.7166244983673096
    },
    {
      "auxiliary_loss_clip": 0.01156457,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.04646659,
      "balance_loss_mlp": 1.01539314,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 3.832966279408726,
      "language_loss": 0.70891762,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.73070854,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.569171190261841
    },
    {
      "auxiliary_loss_clip": 0.0115291,
      "auxiliary_loss_mlp": 0.01022327,
      "balance_loss_clip": 1.04576302,
      "balance_loss_mlp": 1.01516521,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 1.7159205462891784,
      "language_loss": 0.79373091,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81548321,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.6592352390289307
    },
    {
      "auxiliary_loss_clip": 0.01158174,
      "auxiliary_loss_mlp": 0.0102386,
      "balance_loss_clip": 1.04614115,
      "balance_loss_mlp": 1.01626647,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 2.474440721447996,
      "language_loss": 0.74599689,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76781714,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.6222307682037354
    },
    {
      "auxiliary_loss_clip": 0.01153719,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04784966,
      "balance_loss_mlp": 1.02071214,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 1.8423869035276943,
      "language_loss": 0.76026517,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.78208172,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.6489932537078857
    },
    {
      "auxiliary_loss_clip": 0.01152319,
      "auxiliary_loss_mlp": 0.01028225,
      "balance_loss_clip": 1.04510498,
      "balance_loss_mlp": 1.02042258,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 1.7851636051298723,
      "language_loss": 0.81598425,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83778965,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 3.582380533218384
    },
    {
      "auxiliary_loss_clip": 0.0113341,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.0415374,
      "balance_loss_mlp": 1.02256787,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 1.7530623170667146,
      "language_loss": 0.67909825,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.70073736,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 2.6600842475891113
    },
    {
      "auxiliary_loss_clip": 0.01156003,
      "auxiliary_loss_mlp": 0.01024918,
      "balance_loss_clip": 1.0511471,
      "balance_loss_mlp": 1.01709771,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 1.8011236203720618,
      "language_loss": 0.6725992,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69440842,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.8092405796051025
    },
    {
      "auxiliary_loss_clip": 0.01162427,
      "auxiliary_loss_mlp": 0.01024941,
      "balance_loss_clip": 1.04781222,
      "balance_loss_mlp": 1.01770496,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 1.9636255564353546,
      "language_loss": 0.7508899,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77276355,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.5765440464019775
    },
    {
      "auxiliary_loss_clip": 0.01128787,
      "auxiliary_loss_mlp": 0.01029887,
      "balance_loss_clip": 1.04361844,
      "balance_loss_mlp": 1.02255583,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 2.2109035538928845,
      "language_loss": 0.75261045,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77419716,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.810971736907959
    },
    {
      "auxiliary_loss_clip": 0.01160707,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.04837799,
      "balance_loss_mlp": 1.01795697,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 5.309641429039368,
      "language_loss": 0.83120596,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85306537,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 3.109727144241333
    },
    {
      "auxiliary_loss_clip": 0.0115076,
      "auxiliary_loss_mlp": 0.01027737,
      "balance_loss_clip": 1.04165912,
      "balance_loss_mlp": 1.02011371,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.8996816088879713,
      "language_loss": 0.61777008,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.6395551,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.6544976234436035
    },
    {
      "auxiliary_loss_clip": 0.0106205,
      "auxiliary_loss_mlp": 0.01001179,
      "balance_loss_clip": 1.01143944,
      "balance_loss_mlp": 1.00030923,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7518870124628914,
      "language_loss": 0.57106388,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59169614,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 4.251887559890747
    },
    {
      "auxiliary_loss_clip": 0.01162945,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.04847884,
      "balance_loss_mlp": 1.01755047,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.4278384061161664,
      "language_loss": 0.78032362,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80220652,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.5625340938568115
    },
    {
      "auxiliary_loss_clip": 0.01066469,
      "auxiliary_loss_mlp": 0.01001062,
      "balance_loss_clip": 1.01048577,
      "balance_loss_mlp": 1.00020981,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.7926275801481513,
      "language_loss": 0.54226053,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56293583,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.142258882522583
    },
    {
      "auxiliary_loss_clip": 0.01132308,
      "auxiliary_loss_mlp": 0.01027034,
      "balance_loss_clip": 1.04251289,
      "balance_loss_mlp": 1.01923823,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.7181195559348184,
      "language_loss": 0.77421045,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79580384,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 3.730788469314575
    },
    {
      "auxiliary_loss_clip": 0.01172361,
      "auxiliary_loss_mlp": 0.0102542,
      "balance_loss_clip": 1.04741538,
      "balance_loss_mlp": 1.01703668,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.3210392712703927,
      "language_loss": 0.60058057,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62255836,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.6270318031311035
    },
    {
      "auxiliary_loss_clip": 0.01155838,
      "auxiliary_loss_mlp": 0.01023206,
      "balance_loss_clip": 1.04548156,
      "balance_loss_mlp": 1.01570737,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 2.243783997600708,
      "language_loss": 0.74888408,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.77067459,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.6595449447631836
    },
    {
      "auxiliary_loss_clip": 0.0113788,
      "auxiliary_loss_mlp": 0.01028835,
      "balance_loss_clip": 1.0425899,
      "balance_loss_mlp": 1.0213306,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 3.8464982628768367,
      "language_loss": 0.76897907,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.79064626,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.5966930389404297
    },
    {
      "auxiliary_loss_clip": 0.01143885,
      "auxiliary_loss_mlp": 0.0090056,
      "balance_loss_clip": 1.04438436,
      "balance_loss_mlp": 1.00090718,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 5.124458431153852,
      "language_loss": 0.74668527,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.76712972,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.6815972328186035
    },
    {
      "auxiliary_loss_clip": 0.01147462,
      "auxiliary_loss_mlp": 0.01023151,
      "balance_loss_clip": 1.04280317,
      "balance_loss_mlp": 1.01537538,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.9310553453068713,
      "language_loss": 0.72675425,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74846041,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.6821413040161133
    },
    {
      "auxiliary_loss_clip": 0.01155022,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 1.04427719,
      "balance_loss_mlp": 1.01759672,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.6896002687521092,
      "language_loss": 0.7708742,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79268157,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.688671588897705
    },
    {
      "auxiliary_loss_clip": 0.01141172,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.04293561,
      "balance_loss_mlp": 1.01851463,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 11.537102161371399,
      "language_loss": 0.59437191,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61604494,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 3.7336270809173584
    },
    {
      "auxiliary_loss_clip": 0.01137102,
      "auxiliary_loss_mlp": 0.01020666,
      "balance_loss_clip": 1.04136336,
      "balance_loss_mlp": 1.01343644,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.8102052671611604,
      "language_loss": 0.72682273,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.74840045,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.713330030441284
    },
    {
      "auxiliary_loss_clip": 0.01162979,
      "auxiliary_loss_mlp": 0.01022896,
      "balance_loss_clip": 1.04733062,
      "balance_loss_mlp": 1.01544523,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.6362745747010956,
      "language_loss": 0.66129422,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68315303,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.836716651916504
    },
    {
      "auxiliary_loss_clip": 0.0115461,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.04438901,
      "balance_loss_mlp": 1.01908231,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.1728955951504227,
      "language_loss": 0.81111026,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83292639,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.6689934730529785
    },
    {
      "auxiliary_loss_clip": 0.01164895,
      "auxiliary_loss_mlp": 0.01028605,
      "balance_loss_clip": 1.04956365,
      "balance_loss_mlp": 1.02045095,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.1645004432899664,
      "language_loss": 0.70546508,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.72740006,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.6151630878448486
    },
    {
      "auxiliary_loss_clip": 0.01173572,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.04952657,
      "balance_loss_mlp": 1.01739812,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 2.0845003757273135,
      "language_loss": 0.79608458,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81807196,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.501304864883423
    },
    {
      "auxiliary_loss_clip": 0.01153595,
      "auxiliary_loss_mlp": 0.01019188,
      "balance_loss_clip": 1.04774368,
      "balance_loss_mlp": 1.01187181,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.1257398057155426,
      "language_loss": 0.64391452,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66564238,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.7805652618408203
    },
    {
      "auxiliary_loss_clip": 0.01172066,
      "auxiliary_loss_mlp": 0.01021676,
      "balance_loss_clip": 1.04899287,
      "balance_loss_mlp": 1.01434779,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.7693516111578,
      "language_loss": 0.70775723,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.72969466,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.576279640197754
    },
    {
      "auxiliary_loss_clip": 0.0115291,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.04405546,
      "balance_loss_mlp": 1.01986361,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.2690779210071614,
      "language_loss": 0.77721488,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79901296,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.610191583633423
    },
    {
      "auxiliary_loss_clip": 0.01146965,
      "auxiliary_loss_mlp": 0.01022748,
      "balance_loss_clip": 1.04380023,
      "balance_loss_mlp": 1.01570833,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.733636135121952,
      "language_loss": 0.7185443,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.74024141,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.6688244342803955
    },
    {
      "auxiliary_loss_clip": 0.01136509,
      "auxiliary_loss_mlp": 0.01023306,
      "balance_loss_clip": 1.04015338,
      "balance_loss_mlp": 1.01633191,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.9256591909255836,
      "language_loss": 0.7747665,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79636467,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 3.6294748783111572
    },
    {
      "auxiliary_loss_clip": 0.01166261,
      "auxiliary_loss_mlp": 0.00901184,
      "balance_loss_clip": 1.04905498,
      "balance_loss_mlp": 1.00079501,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.7312269940629168,
      "language_loss": 0.8096323,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.83030677,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.7142724990844727
    },
    {
      "auxiliary_loss_clip": 0.01141346,
      "auxiliary_loss_mlp": 0.01018525,
      "balance_loss_clip": 1.04327393,
      "balance_loss_mlp": 1.01214409,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 1.8372778620423857,
      "language_loss": 0.73071194,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75231063,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.588703155517578
    },
    {
      "auxiliary_loss_clip": 0.01169707,
      "auxiliary_loss_mlp": 0.01019993,
      "balance_loss_clip": 1.04763758,
      "balance_loss_mlp": 1.01274812,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.7518401505759464,
      "language_loss": 0.84209275,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86398983,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.5900163650512695
    },
    {
      "auxiliary_loss_clip": 0.01167137,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.05104876,
      "balance_loss_mlp": 1.01924849,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.9669333718198296,
      "language_loss": 0.66954511,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.69148374,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.657956600189209
    },
    {
      "auxiliary_loss_clip": 0.01155069,
      "auxiliary_loss_mlp": 0.01021202,
      "balance_loss_clip": 1.04573035,
      "balance_loss_mlp": 1.01344705,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 1.7057599555662732,
      "language_loss": 0.77541292,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79717565,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.6381754875183105
    },
    {
      "auxiliary_loss_clip": 0.01142522,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.04550135,
      "balance_loss_mlp": 1.01667881,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 2.0617202809100688,
      "language_loss": 0.76143748,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78310257,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.7046263217926025
    },
    {
      "auxiliary_loss_clip": 0.01161649,
      "auxiliary_loss_mlp": 0.01024077,
      "balance_loss_clip": 1.04844522,
      "balance_loss_mlp": 1.01653671,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 2.127903822859995,
      "language_loss": 0.70200294,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.7238602,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 3.584092855453491
    },
    {
      "auxiliary_loss_clip": 0.01156802,
      "auxiliary_loss_mlp": 0.01029164,
      "balance_loss_clip": 1.05033445,
      "balance_loss_mlp": 1.0214777,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 2.0821602298060524,
      "language_loss": 0.823075,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84493464,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.7010338306427
    },
    {
      "auxiliary_loss_clip": 0.01160055,
      "auxiliary_loss_mlp": 0.01021155,
      "balance_loss_clip": 1.04638505,
      "balance_loss_mlp": 1.01325715,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 2.8711638650982145,
      "language_loss": 0.70249152,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.7243036,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.8479971885681152
    },
    {
      "auxiliary_loss_clip": 0.0116214,
      "auxiliary_loss_mlp": 0.0102774,
      "balance_loss_clip": 1.04977751,
      "balance_loss_mlp": 1.02072477,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.9108224218535912,
      "language_loss": 0.74187589,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76377463,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 3.511489152908325
    },
    {
      "auxiliary_loss_clip": 0.01141604,
      "auxiliary_loss_mlp": 0.00900756,
      "balance_loss_clip": 1.04248714,
      "balance_loss_mlp": 1.00080252,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 2.842870801820914,
      "language_loss": 0.63989896,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66032255,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.612133264541626
    },
    {
      "auxiliary_loss_clip": 0.01128947,
      "auxiliary_loss_mlp": 0.01021205,
      "balance_loss_clip": 1.04114127,
      "balance_loss_mlp": 1.01389122,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.5391862481132157,
      "language_loss": 0.74268085,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76418233,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.678292989730835
    },
    {
      "auxiliary_loss_clip": 0.01148972,
      "auxiliary_loss_mlp": 0.01020575,
      "balance_loss_clip": 1.04154348,
      "balance_loss_mlp": 1.01383328,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.6510436825407035,
      "language_loss": 0.8050189,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82671428,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.613175392150879
    },
    {
      "auxiliary_loss_clip": 0.01146728,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.04202414,
      "balance_loss_mlp": 1.02010918,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 2.0231381124327,
      "language_loss": 0.61453921,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.6362763,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.7224314212799072
    },
    {
      "auxiliary_loss_clip": 0.01144408,
      "auxiliary_loss_mlp": 0.01025269,
      "balance_loss_clip": 1.03921556,
      "balance_loss_mlp": 1.01800346,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.9744217283558045,
      "language_loss": 0.8450008,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86669755,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.636540412902832
    },
    {
      "auxiliary_loss_clip": 0.01160216,
      "auxiliary_loss_mlp": 0.00900612,
      "balance_loss_clip": 1.04583216,
      "balance_loss_mlp": 1.00074601,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.845477065805608,
      "language_loss": 0.64814156,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.66874993,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.6314079761505127
    },
    {
      "auxiliary_loss_clip": 0.01139005,
      "auxiliary_loss_mlp": 0.01023008,
      "balance_loss_clip": 1.04020643,
      "balance_loss_mlp": 1.01518798,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.807987113423737,
      "language_loss": 0.68974698,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71136713,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 3.657116174697876
    },
    {
      "auxiliary_loss_clip": 0.01170417,
      "auxiliary_loss_mlp": 0.01023017,
      "balance_loss_clip": 1.04950547,
      "balance_loss_mlp": 1.01565599,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 2.1300106296251746,
      "language_loss": 0.70610154,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72803587,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.5664539337158203
    },
    {
      "auxiliary_loss_clip": 0.01159664,
      "auxiliary_loss_mlp": 0.0102561,
      "balance_loss_clip": 1.04591084,
      "balance_loss_mlp": 1.0183084,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 2.2241961756394555,
      "language_loss": 0.72241706,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74426985,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.6467812061309814
    },
    {
      "auxiliary_loss_clip": 0.01159251,
      "auxiliary_loss_mlp": 0.00901302,
      "balance_loss_clip": 1.04538393,
      "balance_loss_mlp": 1.00077176,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 2.1942212701315262,
      "language_loss": 0.86740637,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.88801181,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.5348317623138428
    },
    {
      "auxiliary_loss_clip": 0.01120583,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.03511786,
      "balance_loss_mlp": 1.02065325,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.501224056158754,
      "language_loss": 0.75061572,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.77210605,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.7414751052856445
    },
    {
      "auxiliary_loss_clip": 0.01157743,
      "auxiliary_loss_mlp": 0.01029271,
      "balance_loss_clip": 1.04898453,
      "balance_loss_mlp": 1.02219307,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 2.3980004844499976,
      "language_loss": 0.74284494,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76471508,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.645627021789551
    },
    {
      "auxiliary_loss_clip": 0.01060647,
      "auxiliary_loss_mlp": 0.01001428,
      "balance_loss_clip": 1.00931811,
      "balance_loss_mlp": 1.00058138,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7587831850802113,
      "language_loss": 0.56283605,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58345687,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.1153695583343506
    },
    {
      "auxiliary_loss_clip": 0.01162565,
      "auxiliary_loss_mlp": 0.01029609,
      "balance_loss_clip": 1.04906881,
      "balance_loss_mlp": 1.02268565,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 2.443474161172802,
      "language_loss": 0.78210652,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80402827,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.6185195446014404
    },
    {
      "auxiliary_loss_clip": 0.01173142,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.05051327,
      "balance_loss_mlp": 1.02149451,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.379425041395023,
      "language_loss": 0.67732036,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.69934303,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.6467154026031494
    },
    {
      "auxiliary_loss_clip": 0.0116164,
      "auxiliary_loss_mlp": 0.01031146,
      "balance_loss_clip": 1.0474484,
      "balance_loss_mlp": 1.02340913,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.8112651497049603,
      "language_loss": 0.76653826,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.7884661,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 3.525322198867798
    },
    {
      "auxiliary_loss_clip": 0.01144985,
      "auxiliary_loss_mlp": 0.01020786,
      "balance_loss_clip": 1.04125941,
      "balance_loss_mlp": 1.01315665,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.6103086392278114,
      "language_loss": 0.70297694,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72463465,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.7320308685302734
    },
    {
      "auxiliary_loss_clip": 0.01174846,
      "auxiliary_loss_mlp": 0.01021026,
      "balance_loss_clip": 1.05110168,
      "balance_loss_mlp": 1.01379037,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.6176017898865824,
      "language_loss": 0.8223666,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84432536,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.598461389541626
    },
    {
      "auxiliary_loss_clip": 0.01148206,
      "auxiliary_loss_mlp": 0.01027342,
      "balance_loss_clip": 1.04728675,
      "balance_loss_mlp": 1.01930714,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.329962111013117,
      "language_loss": 0.74513435,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.76688981,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.5652918815612793
    },
    {
      "auxiliary_loss_clip": 0.01120468,
      "auxiliary_loss_mlp": 0.00900684,
      "balance_loss_clip": 1.03966975,
      "balance_loss_mlp": 1.00072408,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 3.4178687166112063,
      "language_loss": 0.78420937,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80442089,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.833150863647461
    },
    {
      "auxiliary_loss_clip": 0.01134533,
      "auxiliary_loss_mlp": 0.01022668,
      "balance_loss_clip": 1.04315495,
      "balance_loss_mlp": 1.0154829,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 2.1363099656289073,
      "language_loss": 0.83123624,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85280824,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.6755564212799072
    },
    {
      "auxiliary_loss_clip": 0.01171096,
      "auxiliary_loss_mlp": 0.01023128,
      "balance_loss_clip": 1.04939258,
      "balance_loss_mlp": 1.01574266,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 3.1959534318738325,
      "language_loss": 0.77448273,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.79642498,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.5907163619995117
    },
    {
      "auxiliary_loss_clip": 0.01163849,
      "auxiliary_loss_mlp": 0.01027812,
      "balance_loss_clip": 1.0488112,
      "balance_loss_mlp": 1.0200398,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.831621461387813,
      "language_loss": 0.80630708,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82822371,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.6345505714416504
    },
    {
      "auxiliary_loss_clip": 0.01158943,
      "auxiliary_loss_mlp": 0.01025917,
      "balance_loss_clip": 1.05195117,
      "balance_loss_mlp": 1.01805496,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.8150059763858906,
      "language_loss": 0.76588434,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78773296,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 3.5747671127319336
    },
    {
      "auxiliary_loss_clip": 0.01158603,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.04701734,
      "balance_loss_mlp": 1.01861715,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 2.6007839272944464,
      "language_loss": 0.75864148,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.78048831,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.6768765449523926
    },
    {
      "auxiliary_loss_clip": 0.01171989,
      "auxiliary_loss_mlp": 0.00901146,
      "balance_loss_clip": 1.04902232,
      "balance_loss_mlp": 1.00079894,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.6473511312235039,
      "language_loss": 0.80802703,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.82875836,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.5593771934509277
    },
    {
      "auxiliary_loss_clip": 0.01063613,
      "auxiliary_loss_mlp": 0.01001155,
      "balance_loss_clip": 1.01395416,
      "balance_loss_mlp": 1.000368,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8333429493438475,
      "language_loss": 0.60926801,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.62991571,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.9500138759613037
    },
    {
      "auxiliary_loss_clip": 0.01142323,
      "auxiliary_loss_mlp": 0.01026476,
      "balance_loss_clip": 1.04110444,
      "balance_loss_mlp": 1.01822662,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.5234261186709483,
      "language_loss": 0.7915746,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81326258,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.650646209716797
    },
    {
      "auxiliary_loss_clip": 0.01146124,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.04208577,
      "balance_loss_mlp": 1.01786399,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.2816001072886123,
      "language_loss": 0.69050711,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71221685,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.59419584274292
    },
    {
      "auxiliary_loss_clip": 0.01174563,
      "auxiliary_loss_mlp": 0.00901323,
      "balance_loss_clip": 1.05176795,
      "balance_loss_mlp": 1.00097775,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.814199212308706,
      "language_loss": 0.88582939,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.9065882,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.471475124359131
    },
    {
      "auxiliary_loss_clip": 0.01165321,
      "auxiliary_loss_mlp": 0.00901498,
      "balance_loss_clip": 1.04849517,
      "balance_loss_mlp": 1.00088704,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 2.0438384939285617,
      "language_loss": 0.74436927,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76503742,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 2.887505292892456
    },
    {
      "auxiliary_loss_clip": 0.01146034,
      "auxiliary_loss_mlp": 0.010295,
      "balance_loss_clip": 1.04914486,
      "balance_loss_mlp": 1.02193296,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.294949835528115,
      "language_loss": 0.72716868,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.74892402,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.619098424911499
    },
    {
      "auxiliary_loss_clip": 0.01145192,
      "auxiliary_loss_mlp": 0.0102517,
      "balance_loss_clip": 1.04365385,
      "balance_loss_mlp": 1.01793969,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 6.037540004467437,
      "language_loss": 0.71936333,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74106693,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.6393299102783203
    },
    {
      "auxiliary_loss_clip": 0.01168809,
      "auxiliary_loss_mlp": 0.01023016,
      "balance_loss_clip": 1.04812193,
      "balance_loss_mlp": 1.01591444,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 2.1978057530317647,
      "language_loss": 0.64213157,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66404986,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 3.46730899810791
    },
    {
      "auxiliary_loss_clip": 0.01144337,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.04079604,
      "balance_loss_mlp": 1.01913047,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 2.36544673868189,
      "language_loss": 0.70309055,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.724796,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.620197296142578
    },
    {
      "auxiliary_loss_clip": 0.01164635,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.04996657,
      "balance_loss_mlp": 1.01797318,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.741337094019652,
      "language_loss": 0.79878843,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.82069302,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 2.5836713314056396
    },
    {
      "auxiliary_loss_clip": 0.01164605,
      "auxiliary_loss_mlp": 0.00900991,
      "balance_loss_clip": 1.04825306,
      "balance_loss_mlp": 1.00075269,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.705228219819465,
      "language_loss": 0.64800614,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66866207,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.6395883560180664
    },
    {
      "auxiliary_loss_clip": 0.01153463,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 1.047297,
      "balance_loss_mlp": 1.01851404,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.302285673298848,
      "language_loss": 0.71091634,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73271465,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.716505765914917
    },
    {
      "auxiliary_loss_clip": 0.01148017,
      "auxiliary_loss_mlp": 0.01026782,
      "balance_loss_clip": 1.04173875,
      "balance_loss_mlp": 1.01944494,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 2.342112559955451,
      "language_loss": 0.62683851,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64858645,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.6787307262420654
    },
    {
      "auxiliary_loss_clip": 0.01161257,
      "auxiliary_loss_mlp": 0.01025801,
      "balance_loss_clip": 1.04524195,
      "balance_loss_mlp": 1.01765275,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.5331594224151353,
      "language_loss": 0.70742035,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.72929096,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.6215405464172363
    },
    {
      "auxiliary_loss_clip": 0.01155135,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.04522395,
      "balance_loss_mlp": 1.01896191,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 3.779589683924411,
      "language_loss": 0.7538358,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77565068,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.601977825164795
    },
    {
      "auxiliary_loss_clip": 0.01066811,
      "auxiliary_loss_mlp": 0.01001046,
      "balance_loss_clip": 1.01069629,
      "balance_loss_mlp": 1.00025296,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7190821237733395,
      "language_loss": 0.56237495,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58305353,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.167738437652588
    },
    {
      "auxiliary_loss_clip": 0.01170046,
      "auxiliary_loss_mlp": 0.01022092,
      "balance_loss_clip": 1.04721379,
      "balance_loss_mlp": 1.01463532,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.7665062500875204,
      "language_loss": 0.68052328,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.70244455,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 3.56929087638855
    },
    {
      "auxiliary_loss_clip": 0.01174376,
      "auxiliary_loss_mlp": 0.01029049,
      "balance_loss_clip": 1.04938126,
      "balance_loss_mlp": 1.02146399,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 3.878122272848208,
      "language_loss": 0.79995859,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.82199287,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.4926960468292236
    },
    {
      "auxiliary_loss_clip": 0.01171864,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 1.05026376,
      "balance_loss_mlp": 1.01906669,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.5229594707887917,
      "language_loss": 0.70219183,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72417343,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.6319539546966553
    },
    {
      "auxiliary_loss_clip": 0.01149538,
      "auxiliary_loss_mlp": 0.01023582,
      "balance_loss_clip": 1.0417366,
      "balance_loss_mlp": 1.01595235,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.0190643836259126,
      "language_loss": 0.71468496,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.73641622,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.6105549335479736
    },
    {
      "auxiliary_loss_clip": 0.01155456,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.04899406,
      "balance_loss_mlp": 1.02071106,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 2.055275340474845,
      "language_loss": 0.79481876,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81666064,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.6290738582611084
    },
    {
      "auxiliary_loss_clip": 0.01150157,
      "auxiliary_loss_mlp": 0.01024284,
      "balance_loss_clip": 1.047297,
      "balance_loss_mlp": 1.01711929,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.816921867331804,
      "language_loss": 0.76156098,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.78330541,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.6343510150909424
    },
    {
      "auxiliary_loss_clip": 0.01159295,
      "auxiliary_loss_mlp": 0.01023232,
      "balance_loss_clip": 1.04752648,
      "balance_loss_mlp": 1.01565623,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.8886547122974096,
      "language_loss": 0.69936198,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.72118729,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.656006336212158
    },
    {
      "auxiliary_loss_clip": 0.01153375,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.04705465,
      "balance_loss_mlp": 1.01944971,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.6086162729472895,
      "language_loss": 0.66498417,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.68678486,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 3.65165376663208
    },
    {
      "auxiliary_loss_clip": 0.01151419,
      "auxiliary_loss_mlp": 0.01020732,
      "balance_loss_clip": 1.04420471,
      "balance_loss_mlp": 1.01348412,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 1.8718204223794093,
      "language_loss": 0.80285203,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82457352,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.580862522125244
    },
    {
      "auxiliary_loss_clip": 0.01163673,
      "auxiliary_loss_mlp": 0.01028121,
      "balance_loss_clip": 1.04536831,
      "balance_loss_mlp": 1.02053976,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 2.0189274991728587,
      "language_loss": 0.73161185,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75352979,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.6312918663024902
    },
    {
      "auxiliary_loss_clip": 0.01158685,
      "auxiliary_loss_mlp": 0.01027545,
      "balance_loss_clip": 1.0473907,
      "balance_loss_mlp": 1.02064896,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 2.797969091969472,
      "language_loss": 0.72273093,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.7445932,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 3.635071039199829
    },
    {
      "auxiliary_loss_clip": 0.01145448,
      "auxiliary_loss_mlp": 0.01023939,
      "balance_loss_clip": 1.0475378,
      "balance_loss_mlp": 1.01663733,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 2.150769426199375,
      "language_loss": 0.70223677,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.7239306,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.627980947494507
    },
    {
      "auxiliary_loss_clip": 0.01158797,
      "auxiliary_loss_mlp": 0.00901522,
      "balance_loss_clip": 1.045187,
      "balance_loss_mlp": 1.0009377,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.685203518295947,
      "language_loss": 0.84460264,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86520582,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.681600570678711
    },
    {
      "auxiliary_loss_clip": 0.01052985,
      "auxiliary_loss_mlp": 0.01001965,
      "balance_loss_clip": 1.0127542,
      "balance_loss_mlp": 1.0010649,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7348450911045256,
      "language_loss": 0.5594368,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.57998633,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.0662403106689453
    },
    {
      "auxiliary_loss_clip": 0.01132495,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.04358792,
      "balance_loss_mlp": 1.01726747,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 2.0347300257559393,
      "language_loss": 0.80807304,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.82964694,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.754251718521118
    },
    {
      "auxiliary_loss_clip": 0.0113321,
      "auxiliary_loss_mlp": 0.01022229,
      "balance_loss_clip": 1.03985524,
      "balance_loss_mlp": 1.01524019,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 11.265188148034794,
      "language_loss": 0.81648767,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.83804202,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.770199775695801
    },
    {
      "auxiliary_loss_clip": 0.01150852,
      "auxiliary_loss_mlp": 0.01023463,
      "balance_loss_clip": 1.04319334,
      "balance_loss_mlp": 1.0166024,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.590726413638395,
      "language_loss": 0.69814217,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.71988535,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.6158952713012695
    },
    {
      "auxiliary_loss_clip": 0.01158029,
      "auxiliary_loss_mlp": 0.01027304,
      "balance_loss_clip": 1.04782796,
      "balance_loss_mlp": 1.01971602,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.614504895166323,
      "language_loss": 0.64402312,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66587645,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 3.6224617958068848
    },
    {
      "auxiliary_loss_clip": 0.01153031,
      "auxiliary_loss_mlp": 0.00900795,
      "balance_loss_clip": 1.04738212,
      "balance_loss_mlp": 1.00088179,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.7179174543513371,
      "language_loss": 0.80166674,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82220507,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.7168445587158203
    },
    {
      "auxiliary_loss_clip": 0.01145463,
      "auxiliary_loss_mlp": 0.01020644,
      "balance_loss_clip": 1.04646361,
      "balance_loss_mlp": 1.01350963,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 1.98616490328239,
      "language_loss": 0.73823267,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.75989377,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.7063117027282715
    },
    {
      "auxiliary_loss_clip": 0.0105508,
      "auxiliary_loss_mlp": 0.01003764,
      "balance_loss_clip": 1.01268065,
      "balance_loss_mlp": 1.00284624,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8322866662388496,
      "language_loss": 0.6216197,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64220816,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.1546831130981445
    },
    {
      "auxiliary_loss_clip": 0.01171471,
      "auxiliary_loss_mlp": 0.01024966,
      "balance_loss_clip": 1.04797637,
      "balance_loss_mlp": 1.01693416,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.181597242674525,
      "language_loss": 0.75018597,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77215034,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.584110736846924
    },
    {
      "auxiliary_loss_clip": 0.01163661,
      "auxiliary_loss_mlp": 0.01025554,
      "balance_loss_clip": 1.04730725,
      "balance_loss_mlp": 1.01831162,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.5189021644112253,
      "language_loss": 0.83098674,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85287893,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.5584475994110107
    },
    {
      "auxiliary_loss_clip": 0.01152502,
      "auxiliary_loss_mlp": 0.01028577,
      "balance_loss_clip": 1.04689646,
      "balance_loss_mlp": 1.02069128,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 2.156251046629033,
      "language_loss": 0.84460962,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86642039,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.5928709506988525
    },
    {
      "auxiliary_loss_clip": 0.01163926,
      "auxiliary_loss_mlp": 0.0102177,
      "balance_loss_clip": 1.04805863,
      "balance_loss_mlp": 1.01414633,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 3.4933482022369993,
      "language_loss": 0.8627342,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.8845911,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.5960676670074463
    },
    {
      "auxiliary_loss_clip": 0.01118191,
      "auxiliary_loss_mlp": 0.01023786,
      "balance_loss_clip": 1.04041576,
      "balance_loss_mlp": 1.01639497,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 2.8203639314568383,
      "language_loss": 0.75427914,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.7756989,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.885254144668579
    },
    {
      "auxiliary_loss_clip": 0.01135082,
      "auxiliary_loss_mlp": 0.01031661,
      "balance_loss_clip": 1.04670811,
      "balance_loss_mlp": 1.02430034,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.6919362461188059,
      "language_loss": 0.7650674,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78673482,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 3.630856513977051
    },
    {
      "auxiliary_loss_clip": 0.01159823,
      "auxiliary_loss_mlp": 0.01027687,
      "balance_loss_clip": 1.04410434,
      "balance_loss_mlp": 1.01965284,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.883518709317553,
      "language_loss": 0.75471812,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77659321,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.6539316177368164
    },
    {
      "auxiliary_loss_clip": 0.01168956,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.05076826,
      "balance_loss_mlp": 1.01863551,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 2.9960715753062335,
      "language_loss": 0.73570752,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75765604,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.6660351753234863
    },
    {
      "auxiliary_loss_clip": 0.01156359,
      "auxiliary_loss_mlp": 0.01025934,
      "balance_loss_clip": 1.04386663,
      "balance_loss_mlp": 1.01800632,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 2.01822013991149,
      "language_loss": 0.67389119,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69571412,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.778343677520752
    },
    {
      "auxiliary_loss_clip": 0.01164212,
      "auxiliary_loss_mlp": 0.01020757,
      "balance_loss_clip": 1.04999185,
      "balance_loss_mlp": 1.01325297,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.2200418328627487,
      "language_loss": 0.85391629,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87576598,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.6384313106536865
    },
    {
      "auxiliary_loss_clip": 0.01133963,
      "auxiliary_loss_mlp": 0.01020986,
      "balance_loss_clip": 1.04097629,
      "balance_loss_mlp": 1.01433408,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 2.1063457953009896,
      "language_loss": 0.7380327,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.75958216,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.74074125289917
    },
    {
      "auxiliary_loss_clip": 0.01152977,
      "auxiliary_loss_mlp": 0.0090067,
      "balance_loss_clip": 1.04812586,
      "balance_loss_mlp": 1.00076473,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.7565472444623813,
      "language_loss": 0.71422201,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.7347585,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.6808414459228516
    },
    {
      "auxiliary_loss_clip": 0.01160963,
      "auxiliary_loss_mlp": 0.01025545,
      "balance_loss_clip": 1.04568279,
      "balance_loss_mlp": 1.0183742,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 1.9162195834286355,
      "language_loss": 0.82780659,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.84967166,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.6324963569641113
    },
    {
      "auxiliary_loss_clip": 0.01161143,
      "auxiliary_loss_mlp": 0.01031543,
      "balance_loss_clip": 1.04706693,
      "balance_loss_mlp": 1.02396703,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.7875559752845729,
      "language_loss": 0.81063968,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.8325665,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 3.5515215396881104
    },
    {
      "auxiliary_loss_clip": 0.01158434,
      "auxiliary_loss_mlp": 0.01030635,
      "balance_loss_clip": 1.0468384,
      "balance_loss_mlp": 1.02344704,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 2.3877721479125404,
      "language_loss": 0.77175784,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79364854,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.604729652404785
    },
    {
      "auxiliary_loss_clip": 0.0114853,
      "auxiliary_loss_mlp": 0.01028295,
      "balance_loss_clip": 1.04659081,
      "balance_loss_mlp": 1.02045083,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.542028723609041,
      "language_loss": 0.74009424,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76186252,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.629908323287964
    },
    {
      "auxiliary_loss_clip": 0.01121029,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.03819585,
      "balance_loss_mlp": 1.01484966,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 1.9840647922148442,
      "language_loss": 0.72613651,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74757427,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 3.739907741546631
    },
    {
      "auxiliary_loss_clip": 0.01150405,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.04519892,
      "balance_loss_mlp": 1.01644421,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 2.2691416540108045,
      "language_loss": 0.75125825,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77299905,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.7987821102142334
    },
    {
      "auxiliary_loss_clip": 0.0113872,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.04283297,
      "balance_loss_mlp": 1.01820922,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 3.207969254163153,
      "language_loss": 0.78326654,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80490553,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.7122645378112793
    },
    {
      "auxiliary_loss_clip": 0.01174085,
      "auxiliary_loss_mlp": 0.00901841,
      "balance_loss_clip": 1.04752707,
      "balance_loss_mlp": 1.00083351,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.380327297288675,
      "language_loss": 0.79931128,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82007051,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.6641533374786377
    },
    {
      "auxiliary_loss_clip": 0.01152313,
      "auxiliary_loss_mlp": 0.01025674,
      "balance_loss_clip": 1.04492116,
      "balance_loss_mlp": 1.01903439,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.3121638672988025,
      "language_loss": 0.70922875,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.73100865,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.7279586791992188
    },
    {
      "auxiliary_loss_clip": 0.01139447,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.04132819,
      "balance_loss_mlp": 1.01661062,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 3.1648843508832782,
      "language_loss": 0.77472204,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79636085,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.6262717247009277
    },
    {
      "auxiliary_loss_clip": 0.0117183,
      "auxiliary_loss_mlp": 0.01026324,
      "balance_loss_clip": 1.04971623,
      "balance_loss_mlp": 1.01887894,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.685854273609122,
      "language_loss": 0.73711634,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75909793,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 3.414944648742676
    },
    {
      "auxiliary_loss_clip": 0.01070451,
      "auxiliary_loss_mlp": 0.01001144,
      "balance_loss_clip": 1.01078224,
      "balance_loss_mlp": 1.00031519,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8798835182515048,
      "language_loss": 0.64470905,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.665425,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.095867395401001
    },
    {
      "auxiliary_loss_clip": 0.01150427,
      "auxiliary_loss_mlp": 0.01022643,
      "balance_loss_clip": 1.04735124,
      "balance_loss_mlp": 1.01534772,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.149299627906136,
      "language_loss": 0.77311552,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79484618,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.5588290691375732
    },
    {
      "auxiliary_loss_clip": 0.01141405,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.04520702,
      "balance_loss_mlp": 1.01879692,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.5040535618707567,
      "language_loss": 0.84038424,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86205947,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.7370641231536865
    },
    {
      "auxiliary_loss_clip": 0.0117471,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.04904199,
      "balance_loss_mlp": 1.01733112,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 1.9157303740571918,
      "language_loss": 0.66794443,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.68994319,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.673832654953003
    },
    {
      "auxiliary_loss_clip": 0.01153043,
      "auxiliary_loss_mlp": 0.01022574,
      "balance_loss_clip": 1.04364491,
      "balance_loss_mlp": 1.01526606,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 2.9929026216862518,
      "language_loss": 0.68657601,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70833218,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.6091878414154053
    },
    {
      "auxiliary_loss_clip": 0.01143381,
      "auxiliary_loss_mlp": 0.01023121,
      "balance_loss_clip": 1.0411576,
      "balance_loss_mlp": 1.0161562,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 2.181306537386643,
      "language_loss": 0.7376461,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75931114,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.741075038909912
    },
    {
      "auxiliary_loss_clip": 0.0115902,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.0473727,
      "balance_loss_mlp": 1.02050757,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.241813287927993,
      "language_loss": 0.7213884,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.74326289,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.582005023956299
    },
    {
      "auxiliary_loss_clip": 0.01162525,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04605031,
      "balance_loss_mlp": 1.01796532,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 1.9954106104809968,
      "language_loss": 0.78107166,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80294907,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.6587538719177246
    },
    {
      "auxiliary_loss_clip": 0.01147468,
      "auxiliary_loss_mlp": 0.01024277,
      "balance_loss_clip": 1.04438496,
      "balance_loss_mlp": 1.01728237,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 1.7905122325084193,
      "language_loss": 0.69002783,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71174526,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 3.5187923908233643
    },
    {
      "auxiliary_loss_clip": 0.01155538,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.04690301,
      "balance_loss_mlp": 1.0176686,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 2.382986762372759,
      "language_loss": 0.74700695,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.76881069,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.594954252243042
    },
    {
      "auxiliary_loss_clip": 0.01139755,
      "auxiliary_loss_mlp": 0.01025744,
      "balance_loss_clip": 1.04595923,
      "balance_loss_mlp": 1.01841295,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 2.641148103472205,
      "language_loss": 0.78926295,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.81091797,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.6607511043548584
    },
    {
      "auxiliary_loss_clip": 0.01148661,
      "auxiliary_loss_mlp": 0.01024287,
      "balance_loss_clip": 1.04237688,
      "balance_loss_mlp": 1.01674151,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.287139807379452,
      "language_loss": 0.78524685,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80697632,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.6095423698425293
    },
    {
      "auxiliary_loss_clip": 0.01165344,
      "auxiliary_loss_mlp": 0.00901073,
      "balance_loss_clip": 1.04585338,
      "balance_loss_mlp": 1.00082636,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 2.0713027562809225,
      "language_loss": 0.7357831,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.75644726,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.584434986114502
    },
    {
      "auxiliary_loss_clip": 0.01140321,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.04923868,
      "balance_loss_mlp": 1.02159882,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 1.8626299700327584,
      "language_loss": 0.8111645,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83286417,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.6781656742095947
    },
    {
      "auxiliary_loss_clip": 0.01130841,
      "auxiliary_loss_mlp": 0.01023495,
      "balance_loss_clip": 1.04119897,
      "balance_loss_mlp": 1.01595473,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.025721561666389,
      "language_loss": 0.81608164,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83762503,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.790771007537842
    },
    {
      "auxiliary_loss_clip": 0.01140235,
      "auxiliary_loss_mlp": 0.01030049,
      "balance_loss_clip": 1.04509616,
      "balance_loss_mlp": 1.02213299,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 4.618191554642859,
      "language_loss": 0.77540451,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79710734,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.723806858062744
    },
    {
      "auxiliary_loss_clip": 0.01060855,
      "auxiliary_loss_mlp": 0.01001357,
      "balance_loss_clip": 1.01066279,
      "balance_loss_mlp": 1.00062966,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7368735394051604,
      "language_loss": 0.53486681,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55548894,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 4.206876993179321
    },
    {
      "auxiliary_loss_clip": 0.01159293,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 1.04765892,
      "balance_loss_mlp": 1.01711333,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 1.93345328871415,
      "language_loss": 0.75741798,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77925444,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.618239402770996
    },
    {
      "auxiliary_loss_clip": 0.01155013,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04478741,
      "balance_loss_mlp": 1.01957917,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 2.362408278007662,
      "language_loss": 0.76619351,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.7880193,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.7008049488067627
    },
    {
      "auxiliary_loss_clip": 0.01156654,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04768372,
      "balance_loss_mlp": 1.02117538,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 2.2377543609097557,
      "language_loss": 0.68777263,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.70962042,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.8978869915008545
    },
    {
      "auxiliary_loss_clip": 0.01137723,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.04384863,
      "balance_loss_mlp": 1.0210402,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 2.186284441103715,
      "language_loss": 0.64328873,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66495126,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 3.726775646209717
    },
    {
      "auxiliary_loss_clip": 0.01169253,
      "auxiliary_loss_mlp": 0.01024779,
      "balance_loss_clip": 1.04894483,
      "balance_loss_mlp": 1.01778758,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.5261549846633846,
      "language_loss": 0.81076956,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.83270991,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.6299169063568115
    },
    {
      "auxiliary_loss_clip": 0.01171515,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.0498116,
      "balance_loss_mlp": 1.02130747,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.018298123939208,
      "language_loss": 0.7884196,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.8104223,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.540971040725708
    },
    {
      "auxiliary_loss_clip": 0.01131379,
      "auxiliary_loss_mlp": 0.01021284,
      "balance_loss_clip": 1.04253995,
      "balance_loss_mlp": 1.01419663,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.8999814898231195,
      "language_loss": 0.73826849,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.75979519,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.700181007385254
    },
    {
      "auxiliary_loss_clip": 0.01144286,
      "auxiliary_loss_mlp": 0.00900818,
      "balance_loss_clip": 1.04168463,
      "balance_loss_mlp": 1.00066471,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 1.885133805518834,
      "language_loss": 0.70590788,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72635895,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.6332974433898926
    },
    {
      "auxiliary_loss_clip": 0.01152886,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.04554629,
      "balance_loss_mlp": 1.02047086,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.799169488763594,
      "language_loss": 0.66864574,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.69045496,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.655756950378418
    },
    {
      "auxiliary_loss_clip": 0.01155955,
      "auxiliary_loss_mlp": 0.01023339,
      "balance_loss_clip": 1.04724026,
      "balance_loss_mlp": 1.01602554,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.34354704747045,
      "language_loss": 0.75079709,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77259004,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.576115131378174
    },
    {
      "auxiliary_loss_clip": 0.01149305,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 1.04578161,
      "balance_loss_mlp": 1.01685143,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.669670610749416,
      "language_loss": 0.71103257,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73276567,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 3.613792657852173
    },
    {
      "auxiliary_loss_clip": 0.01062231,
      "auxiliary_loss_mlp": 0.01001747,
      "balance_loss_clip": 1.01142418,
      "balance_loss_mlp": 1.00096631,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.837195005124756,
      "language_loss": 0.53314227,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55378205,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.2458362579345703
    },
    {
      "auxiliary_loss_clip": 0.0113851,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.04958117,
      "balance_loss_mlp": 1.0195111,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.5889349895197258,
      "language_loss": 0.77883828,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.80048907,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.7134461402893066
    },
    {
      "auxiliary_loss_clip": 0.01144799,
      "auxiliary_loss_mlp": 0.01024787,
      "balance_loss_clip": 1.04055655,
      "balance_loss_mlp": 1.01718724,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 1.8002358703575723,
      "language_loss": 0.71966171,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.74135756,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.694518804550171
    },
    {
      "auxiliary_loss_clip": 0.01150516,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.04500711,
      "balance_loss_mlp": 1.02038825,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.2725190046912527,
      "language_loss": 0.69744682,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71923357,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.63470721244812
    },
    {
      "auxiliary_loss_clip": 0.01155211,
      "auxiliary_loss_mlp": 0.01024812,
      "balance_loss_clip": 1.04838502,
      "balance_loss_mlp": 1.017344,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.687606386238262,
      "language_loss": 0.81673682,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.8385371,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.8041086196899414
    },
    {
      "auxiliary_loss_clip": 0.01153465,
      "auxiliary_loss_mlp": 0.01023342,
      "balance_loss_clip": 1.0476203,
      "balance_loss_mlp": 1.01616263,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.323690121202952,
      "language_loss": 0.85649854,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87826657,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.609905481338501
    },
    {
      "auxiliary_loss_clip": 0.01071019,
      "auxiliary_loss_mlp": 0.01000743,
      "balance_loss_clip": 1.01093626,
      "balance_loss_mlp": 0.99993289,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7833909941378502,
      "language_loss": 0.61445934,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63517696,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.19497013092041
    },
    {
      "auxiliary_loss_clip": 0.01051075,
      "auxiliary_loss_mlp": 0.01003137,
      "balance_loss_clip": 1.01333475,
      "balance_loss_mlp": 1.002195,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7866914322813179,
      "language_loss": 0.58507609,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60561812,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 4.0521955490112305
    },
    {
      "auxiliary_loss_clip": 0.01164643,
      "auxiliary_loss_mlp": 0.01022222,
      "balance_loss_clip": 1.04873991,
      "balance_loss_mlp": 1.01450312,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.1299676053806214,
      "language_loss": 0.85026699,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87213564,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.524599552154541
    },
    {
      "auxiliary_loss_clip": 0.01149917,
      "auxiliary_loss_mlp": 0.01029249,
      "balance_loss_clip": 1.04535556,
      "balance_loss_mlp": 1.02123785,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.124830031802797,
      "language_loss": 0.79580593,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81759757,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.528083562850952
    },
    {
      "auxiliary_loss_clip": 0.01151015,
      "auxiliary_loss_mlp": 0.01023905,
      "balance_loss_clip": 1.04490161,
      "balance_loss_mlp": 1.0167886,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.4669302150087224,
      "language_loss": 0.798464,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.8202132,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.7082271575927734
    },
    {
      "auxiliary_loss_clip": 0.01141923,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.0426755,
      "balance_loss_mlp": 1.01901984,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.9152999388421832,
      "language_loss": 0.72262335,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74431074,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.746798038482666
    },
    {
      "auxiliary_loss_clip": 0.01149309,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.0463593,
      "balance_loss_mlp": 1.01897383,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.6497036293729146,
      "language_loss": 0.67530465,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.69705617,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.0106139183044434
    },
    {
      "auxiliary_loss_clip": 0.01157744,
      "auxiliary_loss_mlp": 0.01027882,
      "balance_loss_clip": 1.04342008,
      "balance_loss_mlp": 1.02057433,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.6101755410643497,
      "language_loss": 0.7716226,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79347885,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.622246265411377
    },
    {
      "auxiliary_loss_clip": 0.01162522,
      "auxiliary_loss_mlp": 0.01021838,
      "balance_loss_clip": 1.04670215,
      "balance_loss_mlp": 1.01477802,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.9044341707684944,
      "language_loss": 0.86885172,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.89069533,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.611224412918091
    },
    {
      "auxiliary_loss_clip": 0.01163068,
      "auxiliary_loss_mlp": 0.01020727,
      "balance_loss_clip": 1.04821467,
      "balance_loss_mlp": 1.01349652,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 2.5792555419093746,
      "language_loss": 0.78042459,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80226254,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 3.5681302547454834
    },
    {
      "auxiliary_loss_clip": 0.01144679,
      "auxiliary_loss_mlp": 0.01023383,
      "balance_loss_clip": 1.04399586,
      "balance_loss_mlp": 1.01632524,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.79433392390958,
      "language_loss": 0.77946889,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.80114949,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.6472911834716797
    },
    {
      "auxiliary_loss_clip": 0.01155251,
      "auxiliary_loss_mlp": 0.01026091,
      "balance_loss_clip": 1.04550803,
      "balance_loss_mlp": 1.01888442,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 1.9242767681242767,
      "language_loss": 0.67819875,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.70001221,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.781254768371582
    },
    {
      "auxiliary_loss_clip": 0.01152448,
      "auxiliary_loss_mlp": 0.01022432,
      "balance_loss_clip": 1.04309154,
      "balance_loss_mlp": 1.01522589,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.480869636142944,
      "language_loss": 0.64653152,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.6682803,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 3.764620542526245
    },
    {
      "auxiliary_loss_clip": 0.01147434,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.04797006,
      "balance_loss_mlp": 1.01949966,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 1.9190251887481724,
      "language_loss": 0.71678621,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.73852265,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.6990206241607666
    },
    {
      "auxiliary_loss_clip": 0.01139045,
      "auxiliary_loss_mlp": 0.01022642,
      "balance_loss_clip": 1.04360628,
      "balance_loss_mlp": 1.015522,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 2.1752449100890523,
      "language_loss": 0.73850036,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76011717,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.661604642868042
    },
    {
      "auxiliary_loss_clip": 0.01158845,
      "auxiliary_loss_mlp": 0.01029807,
      "balance_loss_clip": 1.04372215,
      "balance_loss_mlp": 1.02214789,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.581736887160486,
      "language_loss": 0.73812538,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76001191,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.6672017574310303
    },
    {
      "auxiliary_loss_clip": 0.01066791,
      "auxiliary_loss_mlp": 0.01000096,
      "balance_loss_clip": 1.0108943,
      "balance_loss_mlp": 0.99929714,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.757416819060545,
      "language_loss": 0.53536379,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55603266,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.1089417934417725
    },
    {
      "auxiliary_loss_clip": 0.0117045,
      "auxiliary_loss_mlp": 0.00901992,
      "balance_loss_clip": 1.04674947,
      "balance_loss_mlp": 1.00074816,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 2.5337854324850557,
      "language_loss": 0.79175413,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.81247854,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.6795310974121094
    },
    {
      "auxiliary_loss_clip": 0.01154397,
      "auxiliary_loss_mlp": 0.00901023,
      "balance_loss_clip": 1.05045855,
      "balance_loss_mlp": 1.00084925,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.0075939582654483,
      "language_loss": 0.80244267,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82299685,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.7311184406280518
    },
    {
      "auxiliary_loss_clip": 0.01161336,
      "auxiliary_loss_mlp": 0.0102742,
      "balance_loss_clip": 1.04496253,
      "balance_loss_mlp": 1.02056861,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 2.2280031546698873,
      "language_loss": 0.85555589,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87744349,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 3.5245189666748047
    },
    {
      "auxiliary_loss_clip": 0.01150514,
      "auxiliary_loss_mlp": 0.01021265,
      "balance_loss_clip": 1.0435611,
      "balance_loss_mlp": 1.0139606,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 2.285066132155342,
      "language_loss": 0.65875649,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68047428,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.698091506958008
    },
    {
      "auxiliary_loss_clip": 0.01114719,
      "auxiliary_loss_mlp": 0.01023599,
      "balance_loss_clip": 1.03673887,
      "balance_loss_mlp": 1.01543283,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 2.0107253488699905,
      "language_loss": 0.73246419,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75384742,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.646700620651245
    },
    {
      "auxiliary_loss_clip": 0.01166811,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.04853785,
      "balance_loss_mlp": 1.01620042,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.6779751257380908,
      "language_loss": 0.73191237,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75381815,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.6674156188964844
    },
    {
      "auxiliary_loss_clip": 0.01148357,
      "auxiliary_loss_mlp": 0.01023973,
      "balance_loss_clip": 1.0457381,
      "balance_loss_mlp": 1.0166595,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 3.800038745053968,
      "language_loss": 0.71373338,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73545671,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.584855079650879
    },
    {
      "auxiliary_loss_clip": 0.01151104,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.04658699,
      "balance_loss_mlp": 1.0178659,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 2.985074994967915,
      "language_loss": 0.69903678,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72079682,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.7925174236297607
    },
    {
      "auxiliary_loss_clip": 0.01160912,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.04808748,
      "balance_loss_mlp": 1.01895463,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.01966495038413,
      "language_loss": 0.73368919,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75556272,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.6418657302856445
    },
    {
      "auxiliary_loss_clip": 0.01133297,
      "auxiliary_loss_mlp": 0.0102872,
      "balance_loss_clip": 1.0432049,
      "balance_loss_mlp": 1.02128077,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.6680053255030505,
      "language_loss": 0.8553741,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87699425,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.66607403755188
    },
    {
      "auxiliary_loss_clip": 0.01162439,
      "auxiliary_loss_mlp": 0.0102479,
      "balance_loss_clip": 1.04711664,
      "balance_loss_mlp": 1.01731229,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 2.0586928160758537,
      "language_loss": 0.72804534,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74991763,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.7232460975646973
    },
    {
      "auxiliary_loss_clip": 0.01145029,
      "auxiliary_loss_mlp": 0.0102166,
      "balance_loss_clip": 1.04493928,
      "balance_loss_mlp": 1.01414394,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.930279036338432,
      "language_loss": 0.86071587,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88238269,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 3.690763473510742
    },
    {
      "auxiliary_loss_clip": 0.01124056,
      "auxiliary_loss_mlp": 0.01030856,
      "balance_loss_clip": 1.04348946,
      "balance_loss_mlp": 1.02411461,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.7113774402920738,
      "language_loss": 0.63658857,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.65813768,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.764787197113037
    },
    {
      "auxiliary_loss_clip": 0.0114588,
      "auxiliary_loss_mlp": 0.01023746,
      "balance_loss_clip": 1.04454279,
      "balance_loss_mlp": 1.01656711,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.4866736828145726,
      "language_loss": 0.85457826,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87627453,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.617668628692627
    },
    {
      "auxiliary_loss_clip": 0.01152239,
      "auxiliary_loss_mlp": 0.01025111,
      "balance_loss_clip": 1.04330194,
      "balance_loss_mlp": 1.01694512,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.5630387834422055,
      "language_loss": 0.79356492,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81533849,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.684173107147217
    },
    {
      "auxiliary_loss_clip": 0.0115178,
      "auxiliary_loss_mlp": 0.01022169,
      "balance_loss_clip": 1.04314089,
      "balance_loss_mlp": 1.01534462,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.8105794300019018,
      "language_loss": 0.87595385,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89769328,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.6897096633911133
    },
    {
      "auxiliary_loss_clip": 0.01155916,
      "auxiliary_loss_mlp": 0.01021563,
      "balance_loss_clip": 1.04531407,
      "balance_loss_mlp": 1.01380801,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.7719480128305927,
      "language_loss": 0.70708799,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72886282,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.7522854804992676
    },
    {
      "auxiliary_loss_clip": 0.01157486,
      "auxiliary_loss_mlp": 0.01023184,
      "balance_loss_clip": 1.04703712,
      "balance_loss_mlp": 1.01554823,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 2.1487225674504393,
      "language_loss": 0.75867748,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.7804842,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.6355090141296387
    },
    {
      "auxiliary_loss_clip": 0.01145187,
      "auxiliary_loss_mlp": 0.01023298,
      "balance_loss_clip": 1.04446483,
      "balance_loss_mlp": 1.0159483,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.6986311675626922,
      "language_loss": 0.7571553,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77884018,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 3.4047584533691406
    },
    {
      "auxiliary_loss_clip": 0.0116267,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 1.0452987,
      "balance_loss_mlp": 1.01718199,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 1.7617148701607377,
      "language_loss": 0.6946525,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71652567,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.5877535343170166
    },
    {
      "auxiliary_loss_clip": 0.01132018,
      "auxiliary_loss_mlp": 0.01023591,
      "balance_loss_clip": 1.04590917,
      "balance_loss_mlp": 1.0160327,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.6096072422262955,
      "language_loss": 0.75938058,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.78093666,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.727421998977661
    },
    {
      "auxiliary_loss_clip": 0.01160826,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.04686856,
      "balance_loss_mlp": 1.01570404,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.89562666002706,
      "language_loss": 0.7194941,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.74132979,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 3.5628833770751953
    },
    {
      "auxiliary_loss_clip": 0.01167754,
      "auxiliary_loss_mlp": 0.0102283,
      "balance_loss_clip": 1.04737866,
      "balance_loss_mlp": 1.015692,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 2.139600603181552,
      "language_loss": 0.74736714,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.76927292,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.6151483058929443
    },
    {
      "auxiliary_loss_clip": 0.01166187,
      "auxiliary_loss_mlp": 0.00901392,
      "balance_loss_clip": 1.04931092,
      "balance_loss_mlp": 1.00086665,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.1595829058112286,
      "language_loss": 0.74860007,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.7692759,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.6059648990631104
    },
    {
      "auxiliary_loss_clip": 0.01148035,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04417312,
      "balance_loss_mlp": 1.02000654,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 1.8453938390481956,
      "language_loss": 0.78586209,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.80761957,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.648949146270752
    },
    {
      "auxiliary_loss_clip": 0.01147984,
      "auxiliary_loss_mlp": 0.00901461,
      "balance_loss_clip": 1.04826164,
      "balance_loss_mlp": 1.00080752,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 2.066136458094593,
      "language_loss": 0.79324377,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.81373817,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.642307996749878
    },
    {
      "auxiliary_loss_clip": 0.01162867,
      "auxiliary_loss_mlp": 0.01022649,
      "balance_loss_clip": 1.04547465,
      "balance_loss_mlp": 1.01571119,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 1.9174607667036712,
      "language_loss": 0.74246466,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76431984,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.576231002807617
    },
    {
      "auxiliary_loss_clip": 0.01160551,
      "auxiliary_loss_mlp": 0.01023711,
      "balance_loss_clip": 1.04868793,
      "balance_loss_mlp": 1.01702678,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.8597370761125982,
      "language_loss": 0.75594008,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.7777828,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.636030912399292
    },
    {
      "auxiliary_loss_clip": 0.01137365,
      "auxiliary_loss_mlp": 0.01028117,
      "balance_loss_clip": 1.04361618,
      "balance_loss_mlp": 1.02115798,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 2.0363573712724867,
      "language_loss": 0.66653603,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68819082,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 3.594653844833374
    },
    {
      "auxiliary_loss_clip": 0.01159945,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04962254,
      "balance_loss_mlp": 1.01836431,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 1.809375157374139,
      "language_loss": 0.94261611,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96446788,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.571995735168457
    },
    {
      "auxiliary_loss_clip": 0.01169969,
      "auxiliary_loss_mlp": 0.01023491,
      "balance_loss_clip": 1.04888153,
      "balance_loss_mlp": 1.01672876,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 3.2825340328696275,
      "language_loss": 0.85028052,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87221515,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.500338315963745
    },
    {
      "auxiliary_loss_clip": 0.01162761,
      "auxiliary_loss_mlp": 0.01022105,
      "balance_loss_clip": 1.04683411,
      "balance_loss_mlp": 1.01514363,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.7204915996562151,
      "language_loss": 0.75685966,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.7787084,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.5332677364349365
    },
    {
      "auxiliary_loss_clip": 0.01164451,
      "auxiliary_loss_mlp": 0.01020754,
      "balance_loss_clip": 1.04867244,
      "balance_loss_mlp": 1.01351488,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.681156841580316,
      "language_loss": 0.70762122,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.72947335,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.559990882873535
    },
    {
      "auxiliary_loss_clip": 0.01139329,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.04258966,
      "balance_loss_mlp": 1.01716805,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 2.0861091235397353,
      "language_loss": 0.74028891,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76192766,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.553159236907959
    },
    {
      "auxiliary_loss_clip": 0.01151616,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.0476017,
      "balance_loss_mlp": 1.01868916,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.91980414109604,
      "language_loss": 0.70704973,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72882515,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.599344253540039
    },
    {
      "auxiliary_loss_clip": 0.01150719,
      "auxiliary_loss_mlp": 0.01021515,
      "balance_loss_clip": 1.04319727,
      "balance_loss_mlp": 1.01388574,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.2955766825341524,
      "language_loss": 0.64625299,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.66797537,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.851186513900757
    },
    {
      "auxiliary_loss_clip": 0.01135554,
      "auxiliary_loss_mlp": 0.01023437,
      "balance_loss_clip": 1.0453546,
      "balance_loss_mlp": 1.01554489,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.9902358408254601,
      "language_loss": 0.84258211,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86417198,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.6438162326812744
    },
    {
      "auxiliary_loss_clip": 0.01171009,
      "auxiliary_loss_mlp": 0.01021165,
      "balance_loss_clip": 1.04899061,
      "balance_loss_mlp": 1.01364625,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 2.551304843215155,
      "language_loss": 0.75936151,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.78128332,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 3.656684637069702
    },
    {
      "auxiliary_loss_clip": 0.01146436,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.0449307,
      "balance_loss_mlp": 1.01777148,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.9575551421130934,
      "language_loss": 0.7336418,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.7553643,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.7203471660614014
    },
    {
      "auxiliary_loss_clip": 0.01165719,
      "auxiliary_loss_mlp": 0.01020205,
      "balance_loss_clip": 1.0516454,
      "balance_loss_mlp": 1.01334167,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 2.2440934806213977,
      "language_loss": 0.78652525,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80838448,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.664031505584717
    },
    {
      "auxiliary_loss_clip": 0.01175511,
      "auxiliary_loss_mlp": 0.01026455,
      "balance_loss_clip": 1.04946625,
      "balance_loss_mlp": 1.01833987,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 2.255417587969433,
      "language_loss": 0.68806612,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.71008575,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.619568347930908
    },
    {
      "auxiliary_loss_clip": 0.01147059,
      "auxiliary_loss_mlp": 0.01020382,
      "balance_loss_clip": 1.04525077,
      "balance_loss_mlp": 1.01246023,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 2.322200350140771,
      "language_loss": 0.78184724,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80352163,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.7064809799194336
    },
    {
      "auxiliary_loss_clip": 0.01147197,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.04714537,
      "balance_loss_mlp": 1.0165863,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.9995442822052183,
      "language_loss": 0.85403776,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87575203,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.7593448162078857
    },
    {
      "auxiliary_loss_clip": 0.01134507,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.04606497,
      "balance_loss_mlp": 1.02033758,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.7257474551732737,
      "language_loss": 0.76221561,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78383499,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.6633152961730957
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.0090117,
      "balance_loss_clip": 1.04898369,
      "balance_loss_mlp": 1.00073314,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.6320666519668792,
      "language_loss": 0.81391299,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.8345443,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.6783692836761475
    },
    {
      "auxiliary_loss_clip": 0.01158942,
      "auxiliary_loss_mlp": 0.01027167,
      "balance_loss_clip": 1.04418647,
      "balance_loss_mlp": 1.01924217,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.9231438331064008,
      "language_loss": 0.56158257,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.58344364,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 3.561589241027832
    },
    {
      "auxiliary_loss_clip": 0.01169204,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.04758775,
      "balance_loss_mlp": 1.0163765,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 1.8458883501861878,
      "language_loss": 0.82818067,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85010958,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.5864946842193604
    },
    {
      "auxiliary_loss_clip": 0.01171155,
      "auxiliary_loss_mlp": 0.01025864,
      "balance_loss_clip": 1.0492754,
      "balance_loss_mlp": 1.01880133,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.8105991987465784,
      "language_loss": 0.84663695,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86860716,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.568626880645752
    },
    {
      "auxiliary_loss_clip": 0.01139135,
      "auxiliary_loss_mlp": 0.01025868,
      "balance_loss_clip": 1.04557681,
      "balance_loss_mlp": 1.01870942,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.412675541497796,
      "language_loss": 0.80596197,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.82761204,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 3.6554877758026123
    },
    {
      "auxiliary_loss_clip": 0.01151379,
      "auxiliary_loss_mlp": 0.01024561,
      "balance_loss_clip": 1.04601312,
      "balance_loss_mlp": 1.01789105,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 2.1350244467639565,
      "language_loss": 0.7970528,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.81881225,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.636234760284424
    },
    {
      "auxiliary_loss_clip": 0.01166739,
      "auxiliary_loss_mlp": 0.00901638,
      "balance_loss_clip": 1.05071127,
      "balance_loss_mlp": 1.00085545,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.271375855525333,
      "language_loss": 0.58680165,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.60748541,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.553100824356079
    },
    {
      "auxiliary_loss_clip": 0.01149038,
      "auxiliary_loss_mlp": 0.01024009,
      "balance_loss_clip": 1.04551625,
      "balance_loss_mlp": 1.01682627,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.741774328742027,
      "language_loss": 0.73726523,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75899571,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.820384979248047
    },
    {
      "auxiliary_loss_clip": 0.01164474,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.05013776,
      "balance_loss_mlp": 1.01689184,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.18824448058309,
      "language_loss": 0.70979369,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.73168343,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.6375372409820557
    },
    {
      "auxiliary_loss_clip": 0.01143557,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.04204321,
      "balance_loss_mlp": 1.02309549,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 2.8149836374683326,
      "language_loss": 0.77973443,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80148441,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.693197011947632
    },
    {
      "auxiliary_loss_clip": 0.01159766,
      "auxiliary_loss_mlp": 0.01023947,
      "balance_loss_clip": 1.04571128,
      "balance_loss_mlp": 1.01669919,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.0696936293889054,
      "language_loss": 0.73677135,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75860846,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.58819580078125
    },
    {
      "auxiliary_loss_clip": 0.01152815,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.04459071,
      "balance_loss_mlp": 1.02064896,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 2.1128603538516475,
      "language_loss": 0.72016239,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74197686,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 3.5859458446502686
    },
    {
      "auxiliary_loss_clip": 0.01164403,
      "auxiliary_loss_mlp": 0.01026939,
      "balance_loss_clip": 1.0459578,
      "balance_loss_mlp": 1.01874912,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.2652088063561315,
      "language_loss": 0.73478532,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75669873,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.6846325397491455
    },
    {
      "auxiliary_loss_clip": 0.01169549,
      "auxiliary_loss_mlp": 0.01026383,
      "balance_loss_clip": 1.04763579,
      "balance_loss_mlp": 1.01876521,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.8893405096766847,
      "language_loss": 0.79354179,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81550109,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.6167118549346924
    },
    {
      "auxiliary_loss_clip": 0.01142683,
      "auxiliary_loss_mlp": 0.01021724,
      "balance_loss_clip": 1.04691982,
      "balance_loss_mlp": 1.01441634,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 6.089049936385002,
      "language_loss": 0.7035954,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72523946,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.724785804748535
    },
    {
      "auxiliary_loss_clip": 0.01168802,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.04886341,
      "balance_loss_mlp": 1.01948905,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.7075919552603844,
      "language_loss": 0.64595091,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.66790688,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.6330580711364746
    },
    {
      "auxiliary_loss_clip": 0.0116243,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.04838872,
      "balance_loss_mlp": 1.01762915,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.824735461077241,
      "language_loss": 0.74249852,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76437294,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.6310412883758545
    },
    {
      "auxiliary_loss_clip": 0.011317,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 1.04475832,
      "balance_loss_mlp": 1.02100801,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.9341581514514457,
      "language_loss": 0.70259237,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72419131,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.6733853816986084
    },
    {
      "auxiliary_loss_clip": 0.01162895,
      "auxiliary_loss_mlp": 0.01022541,
      "balance_loss_clip": 1.04865396,
      "balance_loss_mlp": 1.01553774,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.9070340350935762,
      "language_loss": 0.73608017,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75793445,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.651743173599243
    },
    {
      "auxiliary_loss_clip": 0.01169097,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.05075932,
      "balance_loss_mlp": 1.01623344,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.6482761262818766,
      "language_loss": 0.79148084,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81339914,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.6021342277526855
    },
    {
      "auxiliary_loss_clip": 0.01154149,
      "auxiliary_loss_mlp": 0.00900513,
      "balance_loss_clip": 1.04843962,
      "balance_loss_mlp": 1.00079453,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5930404041434894,
      "language_loss": 0.85048866,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.87103522,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 3.6209466457366943
    },
    {
      "auxiliary_loss_clip": 0.011639,
      "auxiliary_loss_mlp": 0.01024208,
      "balance_loss_clip": 1.04649174,
      "balance_loss_mlp": 1.01660848,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 1.7801303792719332,
      "language_loss": 0.75230384,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77418488,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.604724884033203
    },
    {
      "auxiliary_loss_clip": 0.01153488,
      "auxiliary_loss_mlp": 0.01021765,
      "balance_loss_clip": 1.04358447,
      "balance_loss_mlp": 1.01473761,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.8916402255136746,
      "language_loss": 0.66395783,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68571031,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.662808418273926
    },
    {
      "auxiliary_loss_clip": 0.01129857,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.04401326,
      "balance_loss_mlp": 1.01674461,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 2.6091883938841884,
      "language_loss": 0.78792834,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80947053,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.8807146549224854
    },
    {
      "auxiliary_loss_clip": 0.01169477,
      "auxiliary_loss_mlp": 0.01029968,
      "balance_loss_clip": 1.04690003,
      "balance_loss_mlp": 1.02264214,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.8606815141237119,
      "language_loss": 0.7943126,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81630707,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.930605888366699
    },
    {
      "auxiliary_loss_clip": 0.01164186,
      "auxiliary_loss_mlp": 0.01029266,
      "balance_loss_clip": 1.04589295,
      "balance_loss_mlp": 1.02159488,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 2.2152175084649413,
      "language_loss": 0.79115438,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81308889,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.6412155628204346
    },
    {
      "auxiliary_loss_clip": 0.01136038,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.04122877,
      "balance_loss_mlp": 1.02315521,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.3783982346753985,
      "language_loss": 0.85357106,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.8752414,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.7333199977874756
    },
    {
      "auxiliary_loss_clip": 0.01178028,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.04996991,
      "balance_loss_mlp": 1.01707864,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.3825489184861426,
      "language_loss": 0.81734389,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.83937728,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.6739840507507324
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.05129838,
      "balance_loss_mlp": 1.01686144,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.461641705489253,
      "language_loss": 0.77866447,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.8006326,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 3.5541179180145264
    },
    {
      "auxiliary_loss_clip": 0.01168736,
      "auxiliary_loss_mlp": 0.01027242,
      "balance_loss_clip": 1.04819083,
      "balance_loss_mlp": 1.0202744,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.7795280578024883,
      "language_loss": 0.72457063,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74653041,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.56054425239563
    },
    {
      "auxiliary_loss_clip": 0.01172138,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.04789841,
      "balance_loss_mlp": 1.02233303,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.7670254142691406,
      "language_loss": 0.74357986,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76560068,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 3.5260818004608154
    },
    {
      "auxiliary_loss_clip": 0.01150951,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 1.04437339,
      "balance_loss_mlp": 1.0172596,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.984698850097029,
      "language_loss": 0.90466201,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.9264124,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.63335919380188
    },
    {
      "auxiliary_loss_clip": 0.01056236,
      "auxiliary_loss_mlp": 0.00891016,
      "balance_loss_clip": 1.01123953,
      "balance_loss_mlp": 1.0000633,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7428185280979955,
      "language_loss": 0.56909227,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.58856475,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.257087469100952
    },
    {
      "auxiliary_loss_clip": 0.01114556,
      "auxiliary_loss_mlp": 0.00901502,
      "balance_loss_clip": 1.03784943,
      "balance_loss_mlp": 1.00086522,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 2.0106157490173375,
      "language_loss": 0.80036974,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.82053035,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.7803120613098145
    },
    {
      "auxiliary_loss_clip": 0.01164341,
      "auxiliary_loss_mlp": 0.01024433,
      "balance_loss_clip": 1.04780769,
      "balance_loss_mlp": 1.01694655,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.8075911227741168,
      "language_loss": 0.73670989,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75859761,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.649162530899048
    },
    {
      "auxiliary_loss_clip": 0.01131904,
      "auxiliary_loss_mlp": 0.00901625,
      "balance_loss_clip": 1.04047918,
      "balance_loss_mlp": 1.00087333,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.4964849598121672,
      "language_loss": 0.70629513,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.72663045,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.6574594974517822
    },
    {
      "auxiliary_loss_clip": 0.01136695,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.044806,
      "balance_loss_mlp": 1.0167346,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.565356565568024,
      "language_loss": 0.75209481,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77369976,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.690004587173462
    },
    {
      "auxiliary_loss_clip": 0.01168958,
      "auxiliary_loss_mlp": 0.01021972,
      "balance_loss_clip": 1.0474112,
      "balance_loss_mlp": 1.01469433,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.7805704464233958,
      "language_loss": 0.81768781,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.83959711,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.6100692749023438
    },
    {
      "auxiliary_loss_clip": 0.01138831,
      "auxiliary_loss_mlp": 0.01024095,
      "balance_loss_clip": 1.04449916,
      "balance_loss_mlp": 1.01719606,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 1.856422542706063,
      "language_loss": 0.84177226,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86340153,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 3.5991098880767822
    },
    {
      "auxiliary_loss_clip": 0.01159951,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.04811954,
      "balance_loss_mlp": 1.01780474,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 1.7817135492800207,
      "language_loss": 0.7559967,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77784729,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.637819290161133
    },
    {
      "auxiliary_loss_clip": 0.01165827,
      "auxiliary_loss_mlp": 0.01024957,
      "balance_loss_clip": 1.0473845,
      "balance_loss_mlp": 1.0178014,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.1844299245979184,
      "language_loss": 0.62429416,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64620203,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.568481206893921
    },
    {
      "auxiliary_loss_clip": 0.01169548,
      "auxiliary_loss_mlp": 0.01021699,
      "balance_loss_clip": 1.05105829,
      "balance_loss_mlp": 1.01332486,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.946194715304672,
      "language_loss": 0.76725066,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78916317,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.789722442626953
    },
    {
      "auxiliary_loss_clip": 0.01165274,
      "auxiliary_loss_mlp": 0.01022209,
      "balance_loss_clip": 1.04802752,
      "balance_loss_mlp": 1.01498175,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 2.0747332378469268,
      "language_loss": 0.80753601,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82941085,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.6307144165039062
    },
    {
      "auxiliary_loss_clip": 0.01151849,
      "auxiliary_loss_mlp": 0.01023181,
      "balance_loss_clip": 1.04466033,
      "balance_loss_mlp": 1.01627827,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 2.4467717598315244,
      "language_loss": 0.77477002,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79652023,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.6624844074249268
    },
    {
      "auxiliary_loss_clip": 0.01161852,
      "auxiliary_loss_mlp": 0.01032368,
      "balance_loss_clip": 1.05079567,
      "balance_loss_mlp": 1.02449977,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.674752378333959,
      "language_loss": 0.59962898,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.62157118,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.8014395236968994
    },
    {
      "auxiliary_loss_clip": 0.01158358,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.04682446,
      "balance_loss_mlp": 1.01606703,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 1.7801329532977919,
      "language_loss": 0.6192528,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.64106888,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.7011945247650146
    },
    {
      "auxiliary_loss_clip": 0.0115078,
      "auxiliary_loss_mlp": 0.01024969,
      "balance_loss_clip": 1.04414153,
      "balance_loss_mlp": 1.01730967,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.049471358941488,
      "language_loss": 0.74680996,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.76856744,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.8570539951324463
    },
    {
      "auxiliary_loss_clip": 0.01147134,
      "auxiliary_loss_mlp": 0.01022179,
      "balance_loss_clip": 1.044101,
      "balance_loss_mlp": 1.01479721,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.8463397659405718,
      "language_loss": 0.75403404,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77572715,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 3.608032703399658
    },
    {
      "auxiliary_loss_clip": 0.0115161,
      "auxiliary_loss_mlp": 0.01023003,
      "balance_loss_clip": 1.04413474,
      "balance_loss_mlp": 1.01612139,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.530177888818829,
      "language_loss": 0.85029125,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87203741,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.677819013595581
    },
    {
      "auxiliary_loss_clip": 0.01135572,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.0435251,
      "balance_loss_mlp": 1.01744866,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.70796526473091,
      "language_loss": 0.63607889,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65768152,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.676060438156128
    },
    {
      "auxiliary_loss_clip": 0.01150313,
      "auxiliary_loss_mlp": 0.01025137,
      "balance_loss_clip": 1.0478549,
      "balance_loss_mlp": 1.01784742,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 3.3972618185413,
      "language_loss": 0.71112919,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73288369,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.75235652923584
    },
    {
      "auxiliary_loss_clip": 0.01170967,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.0494144,
      "balance_loss_mlp": 1.01908469,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.9803929062550292,
      "language_loss": 0.79872417,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.82070029,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.5396158695220947
    },
    {
      "auxiliary_loss_clip": 0.01145299,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.04305947,
      "balance_loss_mlp": 1.02280819,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.4396075943732876,
      "language_loss": 0.80565554,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.82741523,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.7946457862854004
    },
    {
      "auxiliary_loss_clip": 0.0115265,
      "auxiliary_loss_mlp": 0.01030246,
      "balance_loss_clip": 1.04761147,
      "balance_loss_mlp": 1.02312064,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.30903488685866,
      "language_loss": 0.79454058,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81636953,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.6458258628845215
    },
    {
      "auxiliary_loss_clip": 0.01133139,
      "auxiliary_loss_mlp": 0.01019629,
      "balance_loss_clip": 1.0424819,
      "balance_loss_mlp": 1.01275659,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 1.9220245133690208,
      "language_loss": 0.82269865,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84422636,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 3.579200029373169
    },
    {
      "auxiliary_loss_clip": 0.01145236,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.04497457,
      "balance_loss_mlp": 1.01959229,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 3.393827748110482,
      "language_loss": 0.76740706,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78913391,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.6892755031585693
    },
    {
      "auxiliary_loss_clip": 0.01162279,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.04656971,
      "balance_loss_mlp": 1.02005839,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 9.39822942825664,
      "language_loss": 0.69218326,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71408415,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.5834269523620605
    },
    {
      "auxiliary_loss_clip": 0.01177415,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.05197239,
      "balance_loss_mlp": 1.02139902,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 2.694030531852879,
      "language_loss": 0.73095709,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75302696,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 3.481619358062744
    },
    {
      "auxiliary_loss_clip": 0.01065512,
      "auxiliary_loss_mlp": 0.01000249,
      "balance_loss_clip": 1.01032841,
      "balance_loss_mlp": 0.99944443,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8369744258958546,
      "language_loss": 0.5673629,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58802056,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.0786941051483154
    },
    {
      "auxiliary_loss_clip": 0.01169776,
      "auxiliary_loss_mlp": 0.01025852,
      "balance_loss_clip": 1.04870808,
      "balance_loss_mlp": 1.0185411,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.496057169049304,
      "language_loss": 0.85412019,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87607646,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.562438488006592
    },
    {
      "auxiliary_loss_clip": 0.01165166,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.04768395,
      "balance_loss_mlp": 1.01700628,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 2.230835745599638,
      "language_loss": 0.76577854,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78767395,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6407694816589355
    },
    {
      "auxiliary_loss_clip": 0.0117608,
      "auxiliary_loss_mlp": 0.01029341,
      "balance_loss_clip": 1.05058527,
      "balance_loss_mlp": 1.02096081,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 9.815601317078501,
      "language_loss": 0.82170498,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84375918,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.6044857501983643
    },
    {
      "auxiliary_loss_clip": 0.01152272,
      "auxiliary_loss_mlp": 0.01018438,
      "balance_loss_clip": 1.04176724,
      "balance_loss_mlp": 1.01092172,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 1.932418975951609,
      "language_loss": 0.82310164,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.8448087,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.717679023742676
    },
    {
      "auxiliary_loss_clip": 0.01151926,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.04640341,
      "balance_loss_mlp": 1.0216043,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.8392091762572949,
      "language_loss": 0.80520785,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.8270117,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.653722047805786
    },
    {
      "auxiliary_loss_clip": 0.0113467,
      "auxiliary_loss_mlp": 0.01025728,
      "balance_loss_clip": 1.04284453,
      "balance_loss_mlp": 1.01806331,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.8853154696738281,
      "language_loss": 0.77196616,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79357016,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.6961588859558105
    },
    {
      "auxiliary_loss_clip": 0.01158558,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.04674387,
      "balance_loss_mlp": 1.02052307,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 2.5466618265660284,
      "language_loss": 0.81099379,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83286273,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 3.689760208129883
    },
    {
      "auxiliary_loss_clip": 0.01151685,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 1.04519284,
      "balance_loss_mlp": 1.0158298,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.721989864508504,
      "language_loss": 0.61119044,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63293642,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.7435221672058105
    },
    {
      "auxiliary_loss_clip": 0.01140265,
      "auxiliary_loss_mlp": 0.00900854,
      "balance_loss_clip": 1.04657769,
      "balance_loss_mlp": 1.00072408,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.9515073531583829,
      "language_loss": 0.80340528,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.82381648,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.7571775913238525
    },
    {
      "auxiliary_loss_clip": 0.01148898,
      "auxiliary_loss_mlp": 0.01028358,
      "balance_loss_clip": 1.04783154,
      "balance_loss_mlp": 1.0205915,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.973610451855188,
      "language_loss": 0.7331413,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75491381,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.585240125656128
    },
    {
      "auxiliary_loss_clip": 0.01156505,
      "auxiliary_loss_mlp": 0.01023964,
      "balance_loss_clip": 1.04643929,
      "balance_loss_mlp": 1.01678181,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.732100390152799,
      "language_loss": 0.82217687,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84398156,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.6898326873779297
    },
    {
      "auxiliary_loss_clip": 0.01161696,
      "auxiliary_loss_mlp": 0.01023991,
      "balance_loss_clip": 1.04587436,
      "balance_loss_mlp": 1.01626635,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.5770183013863466,
      "language_loss": 0.76599908,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78785598,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.675820827484131
    },
    {
      "auxiliary_loss_clip": 0.01171983,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.04832602,
      "balance_loss_mlp": 1.0213989,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 2.1051099444765633,
      "language_loss": 0.74002993,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.7620461,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.649216651916504
    },
    {
      "auxiliary_loss_clip": 0.01137131,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.04339981,
      "balance_loss_mlp": 1.01725674,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.8298780441063585,
      "language_loss": 0.73623061,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75785452,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.6475906372070312
    },
    {
      "auxiliary_loss_clip": 0.01138892,
      "auxiliary_loss_mlp": 0.01023065,
      "balance_loss_clip": 1.043504,
      "balance_loss_mlp": 1.01458323,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.226912642340161,
      "language_loss": 0.74402881,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.76564837,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 3.6271867752075195
    },
    {
      "auxiliary_loss_clip": 0.01042568,
      "auxiliary_loss_mlp": 0.01002885,
      "balance_loss_clip": 1.0096724,
      "balance_loss_mlp": 1.00205028,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.7683872139948785,
      "language_loss": 0.5749743,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59542882,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.2135214805603027
    },
    {
      "auxiliary_loss_clip": 0.01121312,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04006147,
      "balance_loss_mlp": 1.01689541,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 2.535725065664903,
      "language_loss": 0.79692209,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81837696,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 2.8376755714416504
    },
    {
      "auxiliary_loss_clip": 0.01139047,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04327464,
      "balance_loss_mlp": 1.01859379,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 1.8639240307902654,
      "language_loss": 0.72465026,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74630296,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.7211272716522217
    },
    {
      "auxiliary_loss_clip": 0.01137957,
      "auxiliary_loss_mlp": 0.01019975,
      "balance_loss_clip": 1.04408646,
      "balance_loss_mlp": 1.01214254,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 1.7881649395139048,
      "language_loss": 0.80453038,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82610977,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.7070882320404053
    },
    {
      "auxiliary_loss_clip": 0.01161663,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.04720616,
      "balance_loss_mlp": 1.0170902,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 2.5067667819577837,
      "language_loss": 0.76305991,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.784922,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.633241653442383
    },
    {
      "auxiliary_loss_clip": 0.01168297,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.04947484,
      "balance_loss_mlp": 1.02276659,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 1.9496588893457738,
      "language_loss": 0.7481091,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.770087,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.692861795425415
    },
    {
      "auxiliary_loss_clip": 0.01160071,
      "auxiliary_loss_mlp": 0.01023365,
      "balance_loss_clip": 1.04392684,
      "balance_loss_mlp": 1.01618004,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 2.0721901021836944,
      "language_loss": 0.77193117,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79376554,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.608842372894287
    },
    {
      "auxiliary_loss_clip": 0.01165188,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.04890466,
      "balance_loss_mlp": 1.02195966,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 2.55229133644853,
      "language_loss": 0.71769798,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.73965043,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 3.5118141174316406
    },
    {
      "auxiliary_loss_clip": 0.01153716,
      "auxiliary_loss_mlp": 0.01024958,
      "balance_loss_clip": 1.0446167,
      "balance_loss_mlp": 1.01742446,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 2.529709399375414,
      "language_loss": 0.70768434,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.72947109,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.5580272674560547
    },
    {
      "auxiliary_loss_clip": 0.01173457,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.04949069,
      "balance_loss_mlp": 1.02080178,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.6439861570264458,
      "language_loss": 0.71560097,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73761946,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.583996057510376
    },
    {
      "auxiliary_loss_clip": 0.01153763,
      "auxiliary_loss_mlp": 0.0102962,
      "balance_loss_clip": 1.04852152,
      "balance_loss_mlp": 1.02216923,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.5466583543737298,
      "language_loss": 0.68257493,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70440876,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 3.6338107585906982
    },
    {
      "auxiliary_loss_clip": 0.01163778,
      "auxiliary_loss_mlp": 0.0103089,
      "balance_loss_clip": 1.04941154,
      "balance_loss_mlp": 1.02311158,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 2.1295651375817384,
      "language_loss": 0.79928994,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82123661,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.597151756286621
    },
    {
      "auxiliary_loss_clip": 0.01141543,
      "auxiliary_loss_mlp": 0.01025746,
      "balance_loss_clip": 1.0455128,
      "balance_loss_mlp": 1.0182358,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.715217311308684,
      "language_loss": 0.75287259,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77454549,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.6346848011016846
    },
    {
      "auxiliary_loss_clip": 0.01171531,
      "auxiliary_loss_mlp": 0.00901232,
      "balance_loss_clip": 1.04983425,
      "balance_loss_mlp": 1.00089622,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.684899440616778,
      "language_loss": 0.71024698,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73097461,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.6364943981170654
    },
    {
      "auxiliary_loss_clip": 0.01165765,
      "auxiliary_loss_mlp": 0.01020757,
      "balance_loss_clip": 1.04913163,
      "balance_loss_mlp": 1.01346457,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.110467226919393,
      "language_loss": 0.61009467,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63195992,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.542264699935913
    },
    {
      "auxiliary_loss_clip": 0.0114529,
      "auxiliary_loss_mlp": 0.01021849,
      "balance_loss_clip": 1.04235303,
      "balance_loss_mlp": 1.01480675,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 3.094063356391563,
      "language_loss": 0.82348841,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84515977,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.7033190727233887
    },
    {
      "auxiliary_loss_clip": 0.01152426,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 1.04584908,
      "balance_loss_mlp": 1.0199945,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.811393742970161,
      "language_loss": 0.8137902,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83558875,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.70170521736145
    },
    {
      "auxiliary_loss_clip": 0.01169835,
      "auxiliary_loss_mlp": 0.01026678,
      "balance_loss_clip": 1.0492785,
      "balance_loss_mlp": 1.01972795,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 2.382503581362194,
      "language_loss": 0.77537149,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79733658,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.6018130779266357
    },
    {
      "auxiliary_loss_clip": 0.01150971,
      "auxiliary_loss_mlp": 0.00901692,
      "balance_loss_clip": 1.04323912,
      "balance_loss_mlp": 1.00083601,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 2.399977103124371,
      "language_loss": 0.70845962,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.72898632,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 3.6153457164764404
    },
    {
      "auxiliary_loss_clip": 0.01157412,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.0446372,
      "balance_loss_mlp": 1.01923823,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.8864584754010323,
      "language_loss": 0.83040857,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.85224706,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.640550136566162
    },
    {
      "auxiliary_loss_clip": 0.01156088,
      "auxiliary_loss_mlp": 0.01023925,
      "balance_loss_clip": 1.04353452,
      "balance_loss_mlp": 1.0163939,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.2931268534907234,
      "language_loss": 0.82416451,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84596461,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.649332284927368
    },
    {
      "auxiliary_loss_clip": 0.01151877,
      "auxiliary_loss_mlp": 0.01023333,
      "balance_loss_clip": 1.0465045,
      "balance_loss_mlp": 1.01651073,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.8642644401796464,
      "language_loss": 0.72233844,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.74409056,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.6099801063537598
    },
    {
      "auxiliary_loss_clip": 0.011522,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.04713809,
      "balance_loss_mlp": 1.01812601,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.7268449191548303,
      "language_loss": 0.72314501,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.7449199,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.675945520401001
    },
    {
      "auxiliary_loss_clip": 0.01150992,
      "auxiliary_loss_mlp": 0.01021631,
      "balance_loss_clip": 1.04438114,
      "balance_loss_mlp": 1.01510406,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.7769964341609445,
      "language_loss": 0.79848444,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.8202107,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.63527250289917
    },
    {
      "auxiliary_loss_clip": 0.01156515,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.0474745,
      "balance_loss_mlp": 1.0212698,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 3.5399298934146612,
      "language_loss": 0.73571783,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75757086,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.917149066925049
    },
    {
      "auxiliary_loss_clip": 0.01167972,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.0509547,
      "balance_loss_mlp": 1.01984739,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 2.1209914099467673,
      "language_loss": 0.67987335,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.70183212,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.625001907348633
    },
    {
      "auxiliary_loss_clip": 0.01169757,
      "auxiliary_loss_mlp": 0.01021397,
      "balance_loss_clip": 1.04688716,
      "balance_loss_mlp": 1.01389265,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.2905385519289143,
      "language_loss": 0.69035035,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.71226192,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 3.466930627822876
    },
    {
      "auxiliary_loss_clip": 0.01158288,
      "auxiliary_loss_mlp": 0.01020195,
      "balance_loss_clip": 1.04931307,
      "balance_loss_mlp": 1.01314688,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.0613616741114944,
      "language_loss": 0.78941107,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81119591,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.694620132446289
    },
    {
      "auxiliary_loss_clip": 0.01156157,
      "auxiliary_loss_mlp": 0.01029433,
      "balance_loss_clip": 1.04470086,
      "balance_loss_mlp": 1.02115941,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.7353709103622568,
      "language_loss": 0.65255427,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67441022,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.5883443355560303
    },
    {
      "auxiliary_loss_clip": 0.01165203,
      "auxiliary_loss_mlp": 0.01026082,
      "balance_loss_clip": 1.04886246,
      "balance_loss_mlp": 1.01832771,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 3.2992434731871088,
      "language_loss": 0.68655241,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70846522,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.614330291748047
    },
    {
      "auxiliary_loss_clip": 0.01156847,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.04622555,
      "balance_loss_mlp": 1.0244236,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 1.972093664841123,
      "language_loss": 0.69922155,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.72111094,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.560100793838501
    },
    {
      "auxiliary_loss_clip": 0.01168208,
      "auxiliary_loss_mlp": 0.00901731,
      "balance_loss_clip": 1.04873657,
      "balance_loss_mlp": 1.00074792,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 1.9315547082458004,
      "language_loss": 0.82608253,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84678191,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.6724438667297363
    },
    {
      "auxiliary_loss_clip": 0.0116364,
      "auxiliary_loss_mlp": 0.01025922,
      "balance_loss_clip": 1.04622555,
      "balance_loss_mlp": 1.01850152,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 2.0813632264361437,
      "language_loss": 0.85693789,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87883341,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.6183090209960938
    },
    {
      "auxiliary_loss_clip": 0.01148297,
      "auxiliary_loss_mlp": 0.01024341,
      "balance_loss_clip": 1.0457598,
      "balance_loss_mlp": 1.01698542,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.310032801246575,
      "language_loss": 0.76408088,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78580725,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 3.538735866546631
    },
    {
      "auxiliary_loss_clip": 0.01136191,
      "auxiliary_loss_mlp": 0.01023821,
      "balance_loss_clip": 1.04810846,
      "balance_loss_mlp": 1.01652503,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 2.248515228724305,
      "language_loss": 0.79800487,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.81960499,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.766242742538452
    },
    {
      "auxiliary_loss_clip": 0.01145576,
      "auxiliary_loss_mlp": 0.0102677,
      "balance_loss_clip": 1.04119706,
      "balance_loss_mlp": 1.01971865,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.5707670939234128,
      "language_loss": 0.75741518,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.77913868,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.685523509979248
    },
    {
      "auxiliary_loss_clip": 0.01162111,
      "auxiliary_loss_mlp": 0.00901454,
      "balance_loss_clip": 1.04707003,
      "balance_loss_mlp": 1.00090408,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.270683633156754,
      "language_loss": 0.73114455,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75178021,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.684680938720703
    },
    {
      "auxiliary_loss_clip": 0.01154784,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.04790807,
      "balance_loss_mlp": 1.019104,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 3.1575596001293853,
      "language_loss": 0.7187345,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74054885,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 3.585505247116089
    },
    {
      "auxiliary_loss_clip": 0.01060822,
      "auxiliary_loss_mlp": 0.01001162,
      "balance_loss_clip": 1.01357412,
      "balance_loss_mlp": 1.00029135,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7332968527420384,
      "language_loss": 0.54732347,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56794327,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.2308359146118164
    },
    {
      "auxiliary_loss_clip": 0.01161276,
      "auxiliary_loss_mlp": 0.01030174,
      "balance_loss_clip": 1.04859471,
      "balance_loss_mlp": 1.02323556,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 1.8557206522600416,
      "language_loss": 0.70977259,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73168707,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.6112239360809326
    },
    {
      "auxiliary_loss_clip": 0.01156089,
      "auxiliary_loss_mlp": 0.01023257,
      "balance_loss_clip": 1.04542565,
      "balance_loss_mlp": 1.01580024,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 2.12919902816535,
      "language_loss": 0.70834088,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.73013437,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.698434591293335
    },
    {
      "auxiliary_loss_clip": 0.01171705,
      "auxiliary_loss_mlp": 0.01023554,
      "balance_loss_clip": 1.04942775,
      "balance_loss_mlp": 1.01657712,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.7406681924005343,
      "language_loss": 0.76949668,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79144919,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.551274061203003
    },
    {
      "auxiliary_loss_clip": 0.01150674,
      "auxiliary_loss_mlp": 0.01024448,
      "balance_loss_clip": 1.0464071,
      "balance_loss_mlp": 1.0173012,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.496470209254046,
      "language_loss": 0.83739763,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85914886,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.597799062728882
    },
    {
      "auxiliary_loss_clip": 0.01160459,
      "auxiliary_loss_mlp": 0.0103082,
      "balance_loss_clip": 1.04727769,
      "balance_loss_mlp": 1.02242815,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 2.1345018523687274,
      "language_loss": 0.63098985,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65290272,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.6798508167266846
    },
    {
      "auxiliary_loss_clip": 0.01161521,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.04490256,
      "balance_loss_mlp": 1.01693678,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.748338646429689,
      "language_loss": 0.69743061,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71928936,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 3.6520419120788574
    },
    {
      "auxiliary_loss_clip": 0.01158599,
      "auxiliary_loss_mlp": 0.01028575,
      "balance_loss_clip": 1.04539895,
      "balance_loss_mlp": 1.02068365,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.4161438307188967,
      "language_loss": 0.78256506,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.8044368,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.6119184494018555
    },
    {
      "auxiliary_loss_clip": 0.01140128,
      "auxiliary_loss_mlp": 0.01028812,
      "balance_loss_clip": 1.04361463,
      "balance_loss_mlp": 1.02063465,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.7168524895296073,
      "language_loss": 0.80464447,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82633388,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.737379789352417
    },
    {
      "auxiliary_loss_clip": 0.01172997,
      "auxiliary_loss_mlp": 0.01028086,
      "balance_loss_clip": 1.04959452,
      "balance_loss_mlp": 1.02039146,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 2.748659493098599,
      "language_loss": 0.78138077,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.80339158,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.5308728218078613
    },
    {
      "auxiliary_loss_clip": 0.01165775,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04774356,
      "balance_loss_mlp": 1.01733899,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.9625068212341523,
      "language_loss": 0.700212,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72211826,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.6672415733337402
    },
    {
      "auxiliary_loss_clip": 0.01146891,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 1.04447222,
      "balance_loss_mlp": 1.01746285,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 3.6175177946578,
      "language_loss": 0.78791583,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80963576,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.6534693241119385
    },
    {
      "auxiliary_loss_clip": 0.01154476,
      "auxiliary_loss_mlp": 0.01024452,
      "balance_loss_clip": 1.04599667,
      "balance_loss_mlp": 1.01665258,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 2.346355186790856,
      "language_loss": 0.831065,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85285425,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.7327375411987305
    },
    {
      "auxiliary_loss_clip": 0.01157158,
      "auxiliary_loss_mlp": 0.01024259,
      "balance_loss_clip": 1.04542065,
      "balance_loss_mlp": 1.01717806,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.7044173196640218,
      "language_loss": 0.70877868,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.73059285,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.6670939922332764
    },
    {
      "auxiliary_loss_clip": 0.01173116,
      "auxiliary_loss_mlp": 0.01024635,
      "balance_loss_clip": 1.0516845,
      "balance_loss_mlp": 1.01740456,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 2.0647878708436367,
      "language_loss": 0.84716171,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86913919,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 3.5222365856170654
    },
    {
      "auxiliary_loss_clip": 0.01148667,
      "auxiliary_loss_mlp": 0.01024524,
      "balance_loss_clip": 1.04527354,
      "balance_loss_mlp": 1.01723719,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 1.804513162934022,
      "language_loss": 0.85429984,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87603176,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.6518030166625977
    },
    {
      "auxiliary_loss_clip": 0.01157301,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 1.04679763,
      "balance_loss_mlp": 1.02200031,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 2.3586006385704845,
      "language_loss": 0.78010696,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.80197465,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.663419246673584
    },
    {
      "auxiliary_loss_clip": 0.01164059,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.04570079,
      "balance_loss_mlp": 1.02122688,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.6837087144458334,
      "language_loss": 0.67239225,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69432229,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.6254143714904785
    },
    {
      "auxiliary_loss_clip": 0.01069589,
      "auxiliary_loss_mlp": 0.01000806,
      "balance_loss_clip": 1.01012528,
      "balance_loss_mlp": 1.00003672,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9768480658970788,
      "language_loss": 0.57694757,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59765148,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.153414487838745
    },
    {
      "auxiliary_loss_clip": 0.01148076,
      "auxiliary_loss_mlp": 0.01031979,
      "balance_loss_clip": 1.04577637,
      "balance_loss_mlp": 1.02393222,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.8825486276986405,
      "language_loss": 0.84803271,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.86983329,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.6186201572418213
    },
    {
      "auxiliary_loss_clip": 0.01145684,
      "auxiliary_loss_mlp": 0.01024683,
      "balance_loss_clip": 1.0437243,
      "balance_loss_mlp": 1.01705909,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.580716270542307,
      "language_loss": 0.79955912,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82126284,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.6619229316711426
    },
    {
      "auxiliary_loss_clip": 0.01150499,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.04645848,
      "balance_loss_mlp": 1.01866698,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 1.9771546439557426,
      "language_loss": 0.75066388,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.7724309,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.6886303424835205
    },
    {
      "auxiliary_loss_clip": 0.01143239,
      "auxiliary_loss_mlp": 0.01024135,
      "balance_loss_clip": 1.04423571,
      "balance_loss_mlp": 1.01709569,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.9282740194186463,
      "language_loss": 0.79445153,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81612527,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 3.645693778991699
    },
    {
      "auxiliary_loss_clip": 0.01172388,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.04991853,
      "balance_loss_mlp": 1.01701689,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.7247736625420047,
      "language_loss": 0.71636766,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.73833287,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.638629198074341
    },
    {
      "auxiliary_loss_clip": 0.01060182,
      "auxiliary_loss_mlp": 0.01000689,
      "balance_loss_clip": 1.01050591,
      "balance_loss_mlp": 0.99982494,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8065014843013124,
      "language_loss": 0.5561958,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57680452,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.0366432666778564
    },
    {
      "auxiliary_loss_clip": 0.01145943,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.04480815,
      "balance_loss_mlp": 1.02199101,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 2.5283050728754786,
      "language_loss": 0.73791623,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.7596699,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 3.6316139698028564
    },
    {
      "auxiliary_loss_clip": 0.01155028,
      "auxiliary_loss_mlp": 0.0090062,
      "balance_loss_clip": 1.04416764,
      "balance_loss_mlp": 1.00084949,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.816608085452774,
      "language_loss": 0.68734574,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.70790225,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.6909079551696777
    },
    {
      "auxiliary_loss_clip": 0.01154125,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.04637194,
      "balance_loss_mlp": 1.0174607,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.0549678462879144,
      "language_loss": 0.79054439,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81232977,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.605103015899658
    },
    {
      "auxiliary_loss_clip": 0.01170646,
      "auxiliary_loss_mlp": 0.01024921,
      "balance_loss_clip": 1.04786491,
      "balance_loss_mlp": 1.01793492,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 1.8638950682993762,
      "language_loss": 0.67803323,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69998896,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.789339303970337
    },
    {
      "auxiliary_loss_clip": 0.01166679,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.05077851,
      "balance_loss_mlp": 1.01855326,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.4994744721985,
      "language_loss": 0.82706106,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84899592,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.533299207687378
    },
    {
      "auxiliary_loss_clip": 0.01169738,
      "auxiliary_loss_mlp": 0.01024674,
      "balance_loss_clip": 1.04934502,
      "balance_loss_mlp": 1.01755762,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.6733437171873418,
      "language_loss": 0.79553533,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81747943,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.6140739917755127
    },
    {
      "auxiliary_loss_clip": 0.01064759,
      "auxiliary_loss_mlp": 0.01002018,
      "balance_loss_clip": 1.01021957,
      "balance_loss_mlp": 1.00119579,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8849403148066852,
      "language_loss": 0.62955511,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.6502229,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.2955987453460693
    },
    {
      "auxiliary_loss_clip": 0.01156872,
      "auxiliary_loss_mlp": 0.01021639,
      "balance_loss_clip": 1.04947555,
      "balance_loss_mlp": 1.01432514,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.2937440530234894,
      "language_loss": 0.74431676,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76610184,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.6106696128845215
    },
    {
      "auxiliary_loss_clip": 0.01153858,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.044644,
      "balance_loss_mlp": 1.01773834,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.5160776715158986,
      "language_loss": 0.71632099,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.73812002,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 3.5684192180633545
    },
    {
      "auxiliary_loss_clip": 0.01153683,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.04732955,
      "balance_loss_mlp": 1.01788712,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.925655336684274,
      "language_loss": 0.72537088,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74715507,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.6050212383270264
    },
    {
      "auxiliary_loss_clip": 0.01122891,
      "auxiliary_loss_mlp": 0.01032602,
      "balance_loss_clip": 1.04059434,
      "balance_loss_mlp": 1.0249784,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.189617783127932,
      "language_loss": 0.71601278,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73756766,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.7263572216033936
    },
    {
      "auxiliary_loss_clip": 0.0116406,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04745078,
      "balance_loss_mlp": 1.01638532,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 2.262861308253326,
      "language_loss": 0.70124054,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.72312367,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.5869204998016357
    },
    {
      "auxiliary_loss_clip": 0.011238,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04191375,
      "balance_loss_mlp": 1.01814675,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 2.0460215486069453,
      "language_loss": 0.65472299,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.67621642,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.838580846786499
    },
    {
      "auxiliary_loss_clip": 0.01166197,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 1.0480057,
      "balance_loss_mlp": 1.01794052,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.5282316396144293,
      "language_loss": 0.74906802,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77098417,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.6002280712127686
    },
    {
      "auxiliary_loss_clip": 0.01169994,
      "auxiliary_loss_mlp": 0.01022296,
      "balance_loss_clip": 1.04813433,
      "balance_loss_mlp": 1.01494384,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.794435480688197,
      "language_loss": 0.78148818,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80341101,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.55763578414917
    },
    {
      "auxiliary_loss_clip": 0.01143418,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.04671502,
      "balance_loss_mlp": 1.01905751,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 2.0072122360336033,
      "language_loss": 0.69152761,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71322501,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.6687252521514893
    },
    {
      "auxiliary_loss_clip": 0.01160824,
      "auxiliary_loss_mlp": 0.01025302,
      "balance_loss_clip": 1.04666328,
      "balance_loss_mlp": 1.0171119,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 2.809050763725788,
      "language_loss": 0.76794744,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78980875,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 3.5508453845977783
    },
    {
      "auxiliary_loss_clip": 0.01151474,
      "auxiliary_loss_mlp": 0.01022067,
      "balance_loss_clip": 1.043715,
      "balance_loss_mlp": 1.01421142,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 2.2583504484836454,
      "language_loss": 0.83211136,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85384679,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.6663968563079834
    },
    {
      "auxiliary_loss_clip": 0.01144646,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.04951096,
      "balance_loss_mlp": 1.02068353,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 3.167067704004047,
      "language_loss": 0.72489381,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.74662459,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.620187997817993
    },
    {
      "auxiliary_loss_clip": 0.01151551,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.04622316,
      "balance_loss_mlp": 1.01746297,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 3.1580055143693992,
      "language_loss": 0.68713188,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70889759,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.8534345626831055
    },
    {
      "auxiliary_loss_clip": 0.01055816,
      "auxiliary_loss_mlp": 0.01001299,
      "balance_loss_clip": 1.01099133,
      "balance_loss_mlp": 1.00039315,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9156025270466506,
      "language_loss": 0.750835,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77140617,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.344602346420288
    },
    {
      "auxiliary_loss_clip": 0.01174555,
      "auxiliary_loss_mlp": 0.01029929,
      "balance_loss_clip": 1.04941499,
      "balance_loss_mlp": 1.02221012,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 1.9936403346024243,
      "language_loss": 0.6883117,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71035659,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.577627658843994
    },
    {
      "auxiliary_loss_clip": 0.01158263,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.04598618,
      "balance_loss_mlp": 1.02036309,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 2.604990899075641,
      "language_loss": 0.76310015,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78496087,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.714628219604492
    },
    {
      "auxiliary_loss_clip": 0.01159878,
      "auxiliary_loss_mlp": 0.01020489,
      "balance_loss_clip": 1.04637694,
      "balance_loss_mlp": 1.01306796,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.9121799990983976,
      "language_loss": 0.77673399,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.79853761,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.6388723850250244
    },
    {
      "auxiliary_loss_clip": 0.01169505,
      "auxiliary_loss_mlp": 0.01021188,
      "balance_loss_clip": 1.047472,
      "balance_loss_mlp": 1.01382089,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 5.842552891635136,
      "language_loss": 0.74972975,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.77163672,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 3.5715737342834473
    },
    {
      "auxiliary_loss_clip": 0.01153641,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.04771996,
      "balance_loss_mlp": 1.02489829,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.5525359623081034,
      "language_loss": 0.72477114,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74663788,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.611361503601074
    },
    {
      "auxiliary_loss_clip": 0.01144566,
      "auxiliary_loss_mlp": 0.00900957,
      "balance_loss_clip": 1.0424583,
      "balance_loss_mlp": 1.00074589,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 2.08476667348005,
      "language_loss": 0.68466675,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70512193,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 3.6971018314361572
    },
    {
      "auxiliary_loss_clip": 0.01154184,
      "auxiliary_loss_mlp": 0.01029081,
      "balance_loss_clip": 1.04495895,
      "balance_loss_mlp": 1.02139187,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 3.1941691136278583,
      "language_loss": 0.6541121,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67594481,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.625232458114624
    },
    {
      "auxiliary_loss_clip": 0.01136949,
      "auxiliary_loss_mlp": 0.0102333,
      "balance_loss_clip": 1.0435307,
      "balance_loss_mlp": 1.01590014,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.852759406471462,
      "language_loss": 0.82661313,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84821594,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.6859562397003174
    },
    {
      "auxiliary_loss_clip": 0.01138616,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.04160595,
      "balance_loss_mlp": 1.01847744,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 2.2255171906852484,
      "language_loss": 0.79676366,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81840962,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.674306631088257
    },
    {
      "auxiliary_loss_clip": 0.01155236,
      "auxiliary_loss_mlp": 0.00900791,
      "balance_loss_clip": 1.04535031,
      "balance_loss_mlp": 1.00081944,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 2.9480177891789054,
      "language_loss": 0.82987988,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85044014,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.7010698318481445
    },
    {
      "auxiliary_loss_clip": 0.01161994,
      "auxiliary_loss_mlp": 0.0102658,
      "balance_loss_clip": 1.04856443,
      "balance_loss_mlp": 1.01930225,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 1.9325944306643998,
      "language_loss": 0.67893654,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70082223,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.612090587615967
    },
    {
      "auxiliary_loss_clip": 0.01159818,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 1.04742837,
      "balance_loss_mlp": 1.01882207,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 5.132814656760846,
      "language_loss": 0.70661235,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.72846979,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.618182420730591
    },
    {
      "auxiliary_loss_clip": 0.01154398,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.04777837,
      "balance_loss_mlp": 1.01760936,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.8177918569172602,
      "language_loss": 0.67960775,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.701401,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.6411890983581543
    },
    {
      "auxiliary_loss_clip": 0.01161254,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.04790258,
      "balance_loss_mlp": 1.0169673,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 1.8405261269142832,
      "language_loss": 0.77204037,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79389441,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.620070457458496
    },
    {
      "auxiliary_loss_clip": 0.01173148,
      "auxiliary_loss_mlp": 0.01023441,
      "balance_loss_clip": 1.04995513,
      "balance_loss_mlp": 1.01574063,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 3.63009979475446,
      "language_loss": 0.72609288,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.7480588,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 3.520629405975342
    },
    {
      "auxiliary_loss_clip": 0.01143599,
      "auxiliary_loss_mlp": 0.01019839,
      "balance_loss_clip": 1.04583788,
      "balance_loss_mlp": 1.01336336,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 2.481804670594149,
      "language_loss": 0.71055579,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.73219013,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.6444358825683594
    },
    {
      "auxiliary_loss_clip": 0.01113493,
      "auxiliary_loss_mlp": 0.01025606,
      "balance_loss_clip": 1.04001439,
      "balance_loss_mlp": 1.01802397,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.9144545811547138,
      "language_loss": 0.77187109,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79326206,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.717491865158081
    },
    {
      "auxiliary_loss_clip": 0.01153935,
      "auxiliary_loss_mlp": 0.0102094,
      "balance_loss_clip": 1.04496837,
      "balance_loss_mlp": 1.01415122,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.6522838681227248,
      "language_loss": 0.72979724,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.75154597,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.636934995651245
    },
    {
      "auxiliary_loss_clip": 0.01173666,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.05115032,
      "balance_loss_mlp": 1.01762176,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.9040076214205957,
      "language_loss": 0.69072616,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71271265,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.5262699127197266
    },
    {
      "auxiliary_loss_clip": 0.01133461,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 1.04881072,
      "balance_loss_mlp": 1.01690793,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.4623531167827353,
      "language_loss": 0.72908437,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.75065976,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.704437017440796
    },
    {
      "auxiliary_loss_clip": 0.01151444,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.04758632,
      "balance_loss_mlp": 1.01966453,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 6.202923692512425,
      "language_loss": 0.73390293,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75568753,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.59602952003479
    },
    {
      "auxiliary_loss_clip": 0.01156932,
      "auxiliary_loss_mlp": 0.00901668,
      "balance_loss_clip": 1.04629552,
      "balance_loss_mlp": 1.00075066,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.7620150289850056,
      "language_loss": 0.71001363,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.73059964,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.708062171936035
    },
    {
      "auxiliary_loss_clip": 0.01132766,
      "auxiliary_loss_mlp": 0.01023217,
      "balance_loss_clip": 1.04436767,
      "balance_loss_mlp": 1.01580191,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 3.5136950365235564,
      "language_loss": 0.73884809,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76040792,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 3.734031915664673
    },
    {
      "auxiliary_loss_clip": 0.01161239,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.04631197,
      "balance_loss_mlp": 1.02048934,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 2.3090598585984505,
      "language_loss": 0.72304142,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74492753,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.6240406036376953
    },
    {
      "auxiliary_loss_clip": 0.01132043,
      "auxiliary_loss_mlp": 0.01023488,
      "balance_loss_clip": 1.04553962,
      "balance_loss_mlp": 1.01642799,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 2.4216494772517443,
      "language_loss": 0.77982384,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.8013792,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.8040730953216553
    },
    {
      "auxiliary_loss_clip": 0.01160127,
      "auxiliary_loss_mlp": 0.00901198,
      "balance_loss_clip": 1.04493284,
      "balance_loss_mlp": 1.00075853,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.4263538920565257,
      "language_loss": 0.77133262,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.79194593,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.5909063816070557
    },
    {
      "auxiliary_loss_clip": 0.0113425,
      "auxiliary_loss_mlp": 0.0102306,
      "balance_loss_clip": 1.04603267,
      "balance_loss_mlp": 1.01625609,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 2.157523258047374,
      "language_loss": 0.67612517,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.69769824,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.800060987472534
    },
    {
      "auxiliary_loss_clip": 0.01151115,
      "auxiliary_loss_mlp": 0.01021023,
      "balance_loss_clip": 1.04422235,
      "balance_loss_mlp": 1.01375079,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.5307129021696033,
      "language_loss": 0.82050073,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.84222209,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.6248114109039307
    },
    {
      "auxiliary_loss_clip": 0.01166467,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.04849839,
      "balance_loss_mlp": 1.01759195,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 3.8118198468514586,
      "language_loss": 0.75463557,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77655101,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.6018054485321045
    },
    {
      "auxiliary_loss_clip": 0.01150468,
      "auxiliary_loss_mlp": 0.01021982,
      "balance_loss_clip": 1.04482007,
      "balance_loss_mlp": 1.01493406,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 1.8430079269452186,
      "language_loss": 0.73472226,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75644672,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.710329055786133
    },
    {
      "auxiliary_loss_clip": 0.01157734,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04786587,
      "balance_loss_mlp": 1.01781929,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 13.493136577493392,
      "language_loss": 0.71857125,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.74040258,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 3.561039686203003
    },
    {
      "auxiliary_loss_clip": 0.01145286,
      "auxiliary_loss_mlp": 0.00900974,
      "balance_loss_clip": 1.04823446,
      "balance_loss_mlp": 1.0007751,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.278212724943358,
      "language_loss": 0.80286694,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.82332957,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.715709686279297
    },
    {
      "auxiliary_loss_clip": 0.01167991,
      "auxiliary_loss_mlp": 0.01023912,
      "balance_loss_clip": 1.04789603,
      "balance_loss_mlp": 1.01702213,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.7891304169624387,
      "language_loss": 0.78215039,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.8040694,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 3.4518544673919678
    },
    {
      "auxiliary_loss_clip": 0.01162313,
      "auxiliary_loss_mlp": 0.01028079,
      "balance_loss_clip": 1.04934609,
      "balance_loss_mlp": 1.02085805,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.6705016394557686,
      "language_loss": 0.85300016,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87490416,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.662161111831665
    },
    {
      "auxiliary_loss_clip": 0.01158419,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.04498124,
      "balance_loss_mlp": 1.02032566,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.7066427310451042,
      "language_loss": 0.75493336,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77679479,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.629140615463257
    },
    {
      "auxiliary_loss_clip": 0.01124169,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.03858399,
      "balance_loss_mlp": 1.02031696,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 3.084106770068324,
      "language_loss": 0.94497365,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96649379,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.78723406791687
    },
    {
      "auxiliary_loss_clip": 0.01162938,
      "auxiliary_loss_mlp": 0.01032275,
      "balance_loss_clip": 1.05073595,
      "balance_loss_mlp": 1.02417469,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.7003776072874321,
      "language_loss": 0.7764926,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79844469,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.645716428756714
    },
    {
      "auxiliary_loss_clip": 0.01163816,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.04663348,
      "balance_loss_mlp": 1.01829612,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.971669347448206,
      "language_loss": 0.7329191,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75481194,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.7428677082061768
    },
    {
      "auxiliary_loss_clip": 0.01064787,
      "auxiliary_loss_mlp": 0.00891071,
      "balance_loss_clip": 1.01073503,
      "balance_loss_mlp": 0.99996608,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.662805861088429,
      "language_loss": 0.53235728,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55191588,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.13421630859375
    },
    {
      "auxiliary_loss_clip": 0.01136912,
      "auxiliary_loss_mlp": 0.01022138,
      "balance_loss_clip": 1.04403734,
      "balance_loss_mlp": 1.01519716,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 1.8547499679691746,
      "language_loss": 0.73550761,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.7570982,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.633141279220581
    },
    {
      "auxiliary_loss_clip": 0.01169697,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.04896832,
      "balance_loss_mlp": 1.01774585,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 2.699838308795897,
      "language_loss": 0.67922413,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.70117384,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.5833628177642822
    },
    {
      "auxiliary_loss_clip": 0.01135806,
      "auxiliary_loss_mlp": 0.01021169,
      "balance_loss_clip": 1.04642701,
      "balance_loss_mlp": 1.01395392,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.7103407680106766,
      "language_loss": 0.79047781,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.8120476,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 3.634002685546875
    },
    {
      "auxiliary_loss_clip": 0.0118079,
      "auxiliary_loss_mlp": 0.01024981,
      "balance_loss_clip": 1.05607212,
      "balance_loss_mlp": 1.01714253,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.9031369122356,
      "language_loss": 0.72614938,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74820709,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.551676034927368
    },
    {
      "auxiliary_loss_clip": 0.01152282,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.04506969,
      "balance_loss_mlp": 1.01942444,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 3.6848813841506436,
      "language_loss": 0.79246867,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81425989,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.590348958969116
    },
    {
      "auxiliary_loss_clip": 0.01149356,
      "auxiliary_loss_mlp": 0.01026048,
      "balance_loss_clip": 1.04424632,
      "balance_loss_mlp": 1.01841307,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.2549369418016045,
      "language_loss": 0.74658543,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76833951,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.640049457550049
    },
    {
      "auxiliary_loss_clip": 0.01142515,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.04375577,
      "balance_loss_mlp": 1.01954246,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 1.9969338161654189,
      "language_loss": 0.62698436,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.648682,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.618504285812378
    },
    {
      "auxiliary_loss_clip": 0.01171783,
      "auxiliary_loss_mlp": 0.01026175,
      "balance_loss_clip": 1.05052841,
      "balance_loss_mlp": 1.0181582,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 1.8551796613360567,
      "language_loss": 0.73055834,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75253797,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.668801784515381
    },
    {
      "auxiliary_loss_clip": 0.01162066,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.04865646,
      "balance_loss_mlp": 1.02148235,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.5505267495463007,
      "language_loss": 0.80530822,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82721305,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.5887463092803955
    },
    {
      "auxiliary_loss_clip": 0.01163253,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.04620445,
      "balance_loss_mlp": 1.01974869,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.838002494115349,
      "language_loss": 0.739779,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.76169199,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 3.5749001502990723
    },
    {
      "auxiliary_loss_clip": 0.01165825,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 1.04804277,
      "balance_loss_mlp": 1.020239,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.2830536426045156,
      "language_loss": 0.65455997,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67649782,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.60355806350708
    },
    {
      "auxiliary_loss_clip": 0.01160082,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.04892015,
      "balance_loss_mlp": 1.01758957,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.9237325024479044,
      "language_loss": 0.70690072,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.72874606,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.7361934185028076
    },
    {
      "auxiliary_loss_clip": 0.010553,
      "auxiliary_loss_mlp": 0.0100072,
      "balance_loss_clip": 1.01015782,
      "balance_loss_mlp": 0.99993914,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8753725523517107,
      "language_loss": 0.57853651,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.59909666,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.413339376449585
    },
    {
      "auxiliary_loss_clip": 0.01144339,
      "auxiliary_loss_mlp": 0.0102693,
      "balance_loss_clip": 1.04095805,
      "balance_loss_mlp": 1.01959848,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 2.336098492798421,
      "language_loss": 0.68819433,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.709907,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 2.6613988876342773
    },
    {
      "auxiliary_loss_clip": 0.01172664,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.04965329,
      "balance_loss_mlp": 1.01960421,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.79652219405416,
      "language_loss": 0.69737661,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.71937406,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.5877130031585693
    },
    {
      "auxiliary_loss_clip": 0.01132022,
      "auxiliary_loss_mlp": 0.01024489,
      "balance_loss_clip": 1.04188824,
      "balance_loss_mlp": 1.01715136,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.9756217380926573,
      "language_loss": 0.79657024,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81813538,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.770397424697876
    },
    {
      "auxiliary_loss_clip": 0.01139289,
      "auxiliary_loss_mlp": 0.01019152,
      "balance_loss_clip": 1.04455733,
      "balance_loss_mlp": 1.01229179,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.9875104575303764,
      "language_loss": 0.69649577,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71808016,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.6560637950897217
    },
    {
      "auxiliary_loss_clip": 0.01157577,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.04434705,
      "balance_loss_mlp": 1.01664817,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 7.887686717797749,
      "language_loss": 0.66399813,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68581504,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.688258171081543
    },
    {
      "auxiliary_loss_clip": 0.01159872,
      "auxiliary_loss_mlp": 0.01025312,
      "balance_loss_clip": 1.04541433,
      "balance_loss_mlp": 1.01782918,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.5182068369933712,
      "language_loss": 0.83983207,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.86168391,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 3.572998285293579
    },
    {
      "auxiliary_loss_clip": 0.0115398,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.04986942,
      "balance_loss_mlp": 1.01902747,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 3.598961823206656,
      "language_loss": 0.86465961,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88646233,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.8765745162963867
    },
    {
      "auxiliary_loss_clip": 0.01128699,
      "auxiliary_loss_mlp": 0.01027422,
      "balance_loss_clip": 1.04032588,
      "balance_loss_mlp": 1.02044261,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 2.068690430912068,
      "language_loss": 0.70121992,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72278118,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 3.664393663406372
    },
    {
      "auxiliary_loss_clip": 0.01147777,
      "auxiliary_loss_mlp": 0.01025801,
      "balance_loss_clip": 1.04520845,
      "balance_loss_mlp": 1.01840138,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.7632149921510218,
      "language_loss": 0.72585601,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74759179,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.6381900310516357
    },
    {
      "auxiliary_loss_clip": 0.01162619,
      "auxiliary_loss_mlp": 0.01019818,
      "balance_loss_clip": 1.0473249,
      "balance_loss_mlp": 1.01302862,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 2.0580733423613413,
      "language_loss": 0.68073827,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70256263,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.6502089500427246
    },
    {
      "auxiliary_loss_clip": 0.01152342,
      "auxiliary_loss_mlp": 0.01022199,
      "balance_loss_clip": 1.04226995,
      "balance_loss_mlp": 1.01496923,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 1.8877952487768324,
      "language_loss": 0.74867833,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.77042377,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.6423404216766357
    },
    {
      "auxiliary_loss_clip": 0.01163791,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.04596484,
      "balance_loss_mlp": 1.01876605,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 4.883253428152069,
      "language_loss": 0.79786253,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.81976581,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.6368863582611084
    },
    {
      "auxiliary_loss_clip": 0.01167701,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04737866,
      "balance_loss_mlp": 1.01986969,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 4.5457058079355805,
      "language_loss": 0.71262515,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73457843,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.626464366912842
    },
    {
      "auxiliary_loss_clip": 0.01062297,
      "auxiliary_loss_mlp": 0.0100123,
      "balance_loss_clip": 1.01130283,
      "balance_loss_mlp": 1.00038958,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7522377859915184,
      "language_loss": 0.60430598,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62494123,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.289952516555786
    },
    {
      "auxiliary_loss_clip": 0.01169488,
      "auxiliary_loss_mlp": 0.01022482,
      "balance_loss_clip": 1.04918349,
      "balance_loss_mlp": 1.01527548,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.5975904302038573,
      "language_loss": 0.73935401,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.76127374,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.576349973678589
    },
    {
      "auxiliary_loss_clip": 0.01163816,
      "auxiliary_loss_mlp": 0.01025081,
      "balance_loss_clip": 1.05077124,
      "balance_loss_mlp": 1.01764226,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.887149658953861,
      "language_loss": 0.80966002,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83154899,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 3.5527822971343994
    },
    {
      "auxiliary_loss_clip": 0.01158007,
      "auxiliary_loss_mlp": 0.01024527,
      "balance_loss_clip": 1.04870248,
      "balance_loss_mlp": 1.01657605,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.7637199945691684,
      "language_loss": 0.69573998,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.7175653,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.9310977458953857
    },
    {
      "auxiliary_loss_clip": 0.0113595,
      "auxiliary_loss_mlp": 0.01025848,
      "balance_loss_clip": 1.04449213,
      "balance_loss_mlp": 1.01862073,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 1.9943852639871695,
      "language_loss": 0.82421023,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84582818,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.7582547664642334
    },
    {
      "auxiliary_loss_clip": 0.01131076,
      "auxiliary_loss_mlp": 0.01022044,
      "balance_loss_clip": 1.0415262,
      "balance_loss_mlp": 1.0145247,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 2.015492770587475,
      "language_loss": 0.7402755,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76180673,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.750873565673828
    },
    {
      "auxiliary_loss_clip": 0.01149459,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.04462624,
      "balance_loss_mlp": 1.01962817,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.425762217568593,
      "language_loss": 0.84300607,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86477351,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.604658365249634
    },
    {
      "auxiliary_loss_clip": 0.01154439,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.04385316,
      "balance_loss_mlp": 1.02132273,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.6895426639952036,
      "language_loss": 0.70192719,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72376364,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.6736788749694824
    },
    {
      "auxiliary_loss_clip": 0.01070744,
      "auxiliary_loss_mlp": 0.01001702,
      "balance_loss_clip": 1.01109505,
      "balance_loss_mlp": 1.00088584,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7213924050213018,
      "language_loss": 0.59239703,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61312151,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.074491262435913
    },
    {
      "auxiliary_loss_clip": 0.01152324,
      "auxiliary_loss_mlp": 0.00901338,
      "balance_loss_clip": 1.04785109,
      "balance_loss_mlp": 1.00081968,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 2.789587086882695,
      "language_loss": 0.70438194,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.7249186,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.704951286315918
    },
    {
      "auxiliary_loss_clip": 0.01162012,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 1.04676223,
      "balance_loss_mlp": 1.0202539,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 4.136034559519948,
      "language_loss": 0.75622249,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77812022,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 3.707030773162842
    },
    {
      "auxiliary_loss_clip": 0.011662,
      "auxiliary_loss_mlp": 0.01024792,
      "balance_loss_clip": 1.04854655,
      "balance_loss_mlp": 1.01784539,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.6110325192924049,
      "language_loss": 0.70888317,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73079306,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.596662998199463
    },
    {
      "auxiliary_loss_clip": 0.01170226,
      "auxiliary_loss_mlp": 0.01024463,
      "balance_loss_clip": 1.04851031,
      "balance_loss_mlp": 1.0174861,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.9770878295802936,
      "language_loss": 0.7262482,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74819505,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.5687360763549805
    },
    {
      "auxiliary_loss_clip": 0.01152213,
      "auxiliary_loss_mlp": 0.01020221,
      "balance_loss_clip": 1.04283857,
      "balance_loss_mlp": 1.01273417,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.772768222936174,
      "language_loss": 0.80321974,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82494408,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.6142280101776123
    },
    {
      "auxiliary_loss_clip": 0.01160659,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.0472399,
      "balance_loss_mlp": 1.02040088,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 1.668812903987082,
      "language_loss": 0.67770272,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.69958574,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.6678364276885986
    },
    {
      "auxiliary_loss_clip": 0.01153124,
      "auxiliary_loss_mlp": 0.01022601,
      "balance_loss_clip": 1.04727113,
      "balance_loss_mlp": 1.01460779,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 2.0037233886081234,
      "language_loss": 0.73019212,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75194937,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.674003839492798
    },
    {
      "auxiliary_loss_clip": 0.01149466,
      "auxiliary_loss_mlp": 0.01026698,
      "balance_loss_clip": 1.04335403,
      "balance_loss_mlp": 1.01912189,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.2643924046256694,
      "language_loss": 0.70073956,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.72250128,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.7030839920043945
    },
    {
      "auxiliary_loss_clip": 0.011224,
      "auxiliary_loss_mlp": 0.01026745,
      "balance_loss_clip": 1.03728127,
      "balance_loss_mlp": 1.01955104,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.9290273368213147,
      "language_loss": 0.82071924,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84221077,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.7799298763275146
    },
    {
      "auxiliary_loss_clip": 0.01140816,
      "auxiliary_loss_mlp": 0.01021629,
      "balance_loss_clip": 1.04384172,
      "balance_loss_mlp": 1.01476288,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.966005380922863,
      "language_loss": 0.65665579,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67828023,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 3.630510091781616
    },
    {
      "auxiliary_loss_clip": 0.01174627,
      "auxiliary_loss_mlp": 0.01025361,
      "balance_loss_clip": 1.05021846,
      "balance_loss_mlp": 1.01748765,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 3.07739460728883,
      "language_loss": 0.83667469,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85867465,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.6116814613342285
    },
    {
      "auxiliary_loss_clip": 0.01159706,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.04564977,
      "balance_loss_mlp": 1.01674271,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.8088891966054288,
      "language_loss": 0.72834319,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.75018072,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 3.531497001647949
    },
    {
      "auxiliary_loss_clip": 0.01108535,
      "auxiliary_loss_mlp": 0.01025232,
      "balance_loss_clip": 1.03789294,
      "balance_loss_mlp": 1.01780534,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 1.9815911959314032,
      "language_loss": 0.74912065,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77045834,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.804877519607544
    },
    {
      "auxiliary_loss_clip": 0.0105312,
      "auxiliary_loss_mlp": 0.01001835,
      "balance_loss_clip": 1.01475644,
      "balance_loss_mlp": 1.0009644,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.8125077652243515,
      "language_loss": 0.58639479,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60694432,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.2210347652435303
    },
    {
      "auxiliary_loss_clip": 0.01150924,
      "auxiliary_loss_mlp": 0.01020403,
      "balance_loss_clip": 1.05020928,
      "balance_loss_mlp": 1.01352477,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.4271591938277017,
      "language_loss": 0.80470753,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82642078,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.6952414512634277
    },
    {
      "auxiliary_loss_clip": 0.01158793,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.04425097,
      "balance_loss_mlp": 1.01745045,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 3.6912116101591046,
      "language_loss": 0.75912488,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.78097355,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.647392511367798
    },
    {
      "auxiliary_loss_clip": 0.01168245,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.04890728,
      "balance_loss_mlp": 1.01759601,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 4.571914354016256,
      "language_loss": 0.65527898,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67721266,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.5370404720306396
    },
    {
      "auxiliary_loss_clip": 0.01169501,
      "auxiliary_loss_mlp": 0.01022505,
      "balance_loss_clip": 1.04614854,
      "balance_loss_mlp": 1.01520348,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.708661382854919,
      "language_loss": 0.7557357,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77765572,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.7783427238464355
    },
    {
      "auxiliary_loss_clip": 0.01164177,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.05064678,
      "balance_loss_mlp": 1.01702309,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 1.8239464973021782,
      "language_loss": 0.80005491,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.8219372,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.593371629714966
    },
    {
      "auxiliary_loss_clip": 0.01135537,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 1.04408169,
      "balance_loss_mlp": 1.02086258,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 3.090857811021435,
      "language_loss": 0.77002501,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79165995,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 3.677950859069824
    },
    {
      "auxiliary_loss_clip": 0.01136378,
      "auxiliary_loss_mlp": 0.01023122,
      "balance_loss_clip": 1.0441463,
      "balance_loss_mlp": 1.01582646,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.5434364835747507,
      "language_loss": 0.80828786,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.82988286,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.6823232173919678
    },
    {
      "auxiliary_loss_clip": 0.01148844,
      "auxiliary_loss_mlp": 0.01023188,
      "balance_loss_clip": 1.04505873,
      "balance_loss_mlp": 1.01598442,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.589898823501307,
      "language_loss": 0.73617291,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.7578932,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.705087661743164
    },
    {
      "auxiliary_loss_clip": 0.01156752,
      "auxiliary_loss_mlp": 0.0102374,
      "balance_loss_clip": 1.04630816,
      "balance_loss_mlp": 1.01634884,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 1.9895024076516343,
      "language_loss": 0.82642996,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84823489,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.6132867336273193
    },
    {
      "auxiliary_loss_clip": 0.01137929,
      "auxiliary_loss_mlp": 0.01021642,
      "balance_loss_clip": 1.04317665,
      "balance_loss_mlp": 1.0146656,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 2.431180573817338,
      "language_loss": 0.81841373,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.84000945,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.6578269004821777
    },
    {
      "auxiliary_loss_clip": 0.01158265,
      "auxiliary_loss_mlp": 0.01020948,
      "balance_loss_clip": 1.04811406,
      "balance_loss_mlp": 1.01334834,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.667123497808245,
      "language_loss": 0.81964183,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.841434,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.7020723819732666
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.0501399,
      "balance_loss_mlp": 1.01743495,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 2.311516398619148,
      "language_loss": 0.79081225,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.81270057,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.5999186038970947
    },
    {
      "auxiliary_loss_clip": 0.01136806,
      "auxiliary_loss_mlp": 0.01022805,
      "balance_loss_clip": 1.04019046,
      "balance_loss_mlp": 1.01506877,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.7410116125615926,
      "language_loss": 0.74055362,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76214975,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.756732225418091
    },
    {
      "auxiliary_loss_clip": 0.01145963,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.04544437,
      "balance_loss_mlp": 1.01851237,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.9354410487189184,
      "language_loss": 0.85033554,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87205565,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.6091485023498535
    },
    {
      "auxiliary_loss_clip": 0.01158876,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.04854846,
      "balance_loss_mlp": 1.02145195,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 2.5582646469694224,
      "language_loss": 0.78182125,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80369389,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 3.5690255165100098
    },
    {
      "auxiliary_loss_clip": 0.01116802,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.03956103,
      "balance_loss_mlp": 1.0178138,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.2838078359124525,
      "language_loss": 0.60719585,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62861854,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.9363298416137695
    },
    {
      "auxiliary_loss_clip": 0.0114772,
      "auxiliary_loss_mlp": 0.0102871,
      "balance_loss_clip": 1.05140924,
      "balance_loss_mlp": 1.02108681,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 3.6382664697353455,
      "language_loss": 0.67708671,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.69885099,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 2.853215456008911
    },
    {
      "auxiliary_loss_clip": 0.01153046,
      "auxiliary_loss_mlp": 0.01025689,
      "balance_loss_clip": 1.04469967,
      "balance_loss_mlp": 1.01870334,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 3.312306783414444,
      "language_loss": 0.7060656,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.72785294,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.635411262512207
    },
    {
      "auxiliary_loss_clip": 0.01149666,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.04692531,
      "balance_loss_mlp": 1.02182269,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 2.224084809301294,
      "language_loss": 0.80285364,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.8246364,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.6774234771728516
    },
    {
      "auxiliary_loss_clip": 0.01138251,
      "auxiliary_loss_mlp": 0.01029817,
      "balance_loss_clip": 1.04416966,
      "balance_loss_mlp": 1.02287579,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 2.105880430144969,
      "language_loss": 0.79215097,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81383163,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.73591685295105
    },
    {
      "auxiliary_loss_clip": 0.01138512,
      "auxiliary_loss_mlp": 0.01028112,
      "balance_loss_clip": 1.04376769,
      "balance_loss_mlp": 1.02068543,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 2.0228374153111552,
      "language_loss": 0.66163063,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68329686,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.687222957611084
    },
    {
      "auxiliary_loss_clip": 0.011421,
      "auxiliary_loss_mlp": 0.01032265,
      "balance_loss_clip": 1.04271913,
      "balance_loss_mlp": 1.02513587,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 2.746842918755959,
      "language_loss": 0.72733355,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.7490772,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 3.7898452281951904
    },
    {
      "auxiliary_loss_clip": 0.01152837,
      "auxiliary_loss_mlp": 0.01025643,
      "balance_loss_clip": 1.04730153,
      "balance_loss_mlp": 1.01853228,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.6729331284944073,
      "language_loss": 0.68779832,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70958316,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.7118337154388428
    },
    {
      "auxiliary_loss_clip": 0.01134299,
      "auxiliary_loss_mlp": 0.01022687,
      "balance_loss_clip": 1.04381394,
      "balance_loss_mlp": 1.01555526,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.826893896731388,
      "language_loss": 0.8179791,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83954901,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 3.648921251296997
    },
    {
      "auxiliary_loss_clip": 0.0114943,
      "auxiliary_loss_mlp": 0.01025515,
      "balance_loss_clip": 1.04731286,
      "balance_loss_mlp": 1.01807666,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 1.9151647454911078,
      "language_loss": 0.77366424,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79541361,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.632624387741089
    },
    {
      "auxiliary_loss_clip": 0.0113958,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.04462934,
      "balance_loss_mlp": 1.01950145,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.6883186725326045,
      "language_loss": 0.79406059,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.8157289,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.759270668029785
    },
    {
      "auxiliary_loss_clip": 0.01169132,
      "auxiliary_loss_mlp": 0.01021639,
      "balance_loss_clip": 1.04664695,
      "balance_loss_mlp": 1.01434922,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 1.8874682427122496,
      "language_loss": 0.66450137,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68640906,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.6438040733337402
    },
    {
      "auxiliary_loss_clip": 0.01169258,
      "auxiliary_loss_mlp": 0.01021973,
      "balance_loss_clip": 1.0499481,
      "balance_loss_mlp": 1.0150466,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.811599876954375,
      "language_loss": 0.79476386,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.8166762,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.5988190174102783
    },
    {
      "auxiliary_loss_clip": 0.01146693,
      "auxiliary_loss_mlp": 0.0102765,
      "balance_loss_clip": 1.04431427,
      "balance_loss_mlp": 1.02004409,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.4363237459261518,
      "language_loss": 0.76075304,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.78249645,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 2.7163469791412354
    },
    {
      "auxiliary_loss_clip": 0.01163988,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04634464,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 3.6272478029249378,
      "language_loss": 0.77491474,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79682267,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.6537024974823
    },
    {
      "auxiliary_loss_clip": 0.01134697,
      "auxiliary_loss_mlp": 0.01022643,
      "balance_loss_clip": 1.04051185,
      "balance_loss_mlp": 1.01442909,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 2.0886097567524784,
      "language_loss": 0.72400159,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74557495,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.6681342124938965
    },
    {
      "auxiliary_loss_clip": 0.01160581,
      "auxiliary_loss_mlp": 0.00901645,
      "balance_loss_clip": 1.04815543,
      "balance_loss_mlp": 1.0008862,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 3.2756821447479743,
      "language_loss": 0.77874374,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.799366,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.5965874195098877
    },
    {
      "auxiliary_loss_clip": 0.0114176,
      "auxiliary_loss_mlp": 0.01020638,
      "balance_loss_clip": 1.04032874,
      "balance_loss_mlp": 1.01381624,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 2.252735118552989,
      "language_loss": 0.80122757,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.8228516,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 3.6403796672821045
    },
    {
      "auxiliary_loss_clip": 0.01156769,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.04445314,
      "balance_loss_mlp": 1.01730251,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.0242841616171767,
      "language_loss": 0.84470689,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86651838,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.6617541313171387
    },
    {
      "auxiliary_loss_clip": 0.01155892,
      "auxiliary_loss_mlp": 0.00901796,
      "balance_loss_clip": 1.04382133,
      "balance_loss_mlp": 1.00083125,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.9131304023352262,
      "language_loss": 0.79228467,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.81286156,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.5919158458709717
    },
    {
      "auxiliary_loss_clip": 0.01154352,
      "auxiliary_loss_mlp": 0.01028869,
      "balance_loss_clip": 1.04818201,
      "balance_loss_mlp": 1.02105188,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.8294635354273363,
      "language_loss": 0.70045829,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72229052,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.6487200260162354
    },
    {
      "auxiliary_loss_clip": 0.01151397,
      "auxiliary_loss_mlp": 0.00900935,
      "balance_loss_clip": 1.04496098,
      "balance_loss_mlp": 1.0007875,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.320382641838976,
      "language_loss": 0.75407904,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77460235,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.60227632522583
    },
    {
      "auxiliary_loss_clip": 0.01155182,
      "auxiliary_loss_mlp": 0.01027568,
      "balance_loss_clip": 1.05084872,
      "balance_loss_mlp": 1.02030563,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 2.453828131449652,
      "language_loss": 0.68142092,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70324844,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.6579537391662598
    },
    {
      "auxiliary_loss_clip": 0.01162425,
      "auxiliary_loss_mlp": 0.01026745,
      "balance_loss_clip": 1.04767096,
      "balance_loss_mlp": 1.01820993,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.6454020994344773,
      "language_loss": 0.76617628,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.788068,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.6175553798675537
    },
    {
      "auxiliary_loss_clip": 0.01170072,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.04941225,
      "balance_loss_mlp": 1.02237904,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.123406346287152,
      "language_loss": 0.65418029,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67617935,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.5489728450775146
    },
    {
      "auxiliary_loss_clip": 0.011669,
      "auxiliary_loss_mlp": 0.01025299,
      "balance_loss_clip": 1.04738128,
      "balance_loss_mlp": 1.01771772,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 3.430132192134336,
      "language_loss": 0.8612988,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88322079,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 3.5029001235961914
    },
    {
      "auxiliary_loss_clip": 0.0116488,
      "auxiliary_loss_mlp": 0.01022048,
      "balance_loss_clip": 1.04729235,
      "balance_loss_mlp": 1.01512814,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.4674784676911967,
      "language_loss": 0.764328,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78619725,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.6657540798187256
    },
    {
      "auxiliary_loss_clip": 0.01156066,
      "auxiliary_loss_mlp": 0.01029072,
      "balance_loss_clip": 1.04704952,
      "balance_loss_mlp": 1.02138329,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 1.8997785761760893,
      "language_loss": 0.77030802,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.79215944,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.683861494064331
    },
    {
      "auxiliary_loss_clip": 0.01157381,
      "auxiliary_loss_mlp": 0.01024348,
      "balance_loss_clip": 1.04646468,
      "balance_loss_mlp": 1.0172497,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 2.0596221133576784,
      "language_loss": 0.85201204,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.87382936,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.6386287212371826
    },
    {
      "auxiliary_loss_clip": 0.01163169,
      "auxiliary_loss_mlp": 0.01024169,
      "balance_loss_clip": 1.04530692,
      "balance_loss_mlp": 1.01704383,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 1.9594919670922095,
      "language_loss": 0.83204013,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.85391355,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.6477553844451904
    },
    {
      "auxiliary_loss_clip": 0.0116023,
      "auxiliary_loss_mlp": 0.01022805,
      "balance_loss_clip": 1.04569459,
      "balance_loss_mlp": 1.01494288,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 2.109698220884859,
      "language_loss": 0.77563655,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79746687,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.604245662689209
    },
    {
      "auxiliary_loss_clip": 0.01145247,
      "auxiliary_loss_mlp": 0.0102277,
      "balance_loss_clip": 1.04573131,
      "balance_loss_mlp": 1.01552498,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 2.021979870201236,
      "language_loss": 0.67463553,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69631577,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.680752754211426
    },
    {
      "auxiliary_loss_clip": 0.01139439,
      "auxiliary_loss_mlp": 0.01027189,
      "balance_loss_clip": 1.04379058,
      "balance_loss_mlp": 1.0202179,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 2.6100732940130755,
      "language_loss": 0.79149538,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81316167,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 3.609889268875122
    },
    {
      "auxiliary_loss_clip": 0.01155965,
      "auxiliary_loss_mlp": 0.01022055,
      "balance_loss_clip": 1.04889572,
      "balance_loss_mlp": 1.01509619,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 2.1223418556539206,
      "language_loss": 0.75651377,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77829397,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.654865264892578
    },
    {
      "auxiliary_loss_clip": 0.01044942,
      "auxiliary_loss_mlp": 0.00891216,
      "balance_loss_clip": 1.03030157,
      "balance_loss_mlp": 0.99999857,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.8347255652347093,
      "language_loss": 0.64569497,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66505647,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 4.504685163497925
    },
    {
      "auxiliary_loss_clip": 0.01164553,
      "auxiliary_loss_mlp": 0.01022882,
      "balance_loss_clip": 1.04580235,
      "balance_loss_mlp": 1.01545537,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 3.0872457175596266,
      "language_loss": 0.77549505,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79736948,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 4.051985502243042
    },
    {
      "auxiliary_loss_clip": 0.01152179,
      "auxiliary_loss_mlp": 0.01023831,
      "balance_loss_clip": 1.04524112,
      "balance_loss_mlp": 1.01675606,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.7302802406418607,
      "language_loss": 0.81142336,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83318347,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.6589417457580566
    },
    {
      "auxiliary_loss_clip": 0.01141327,
      "auxiliary_loss_mlp": 0.01023251,
      "balance_loss_clip": 1.04599369,
      "balance_loss_mlp": 1.0158062,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 1.6753422256771233,
      "language_loss": 0.7732811,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79492688,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 2.6871302127838135
    },
    {
      "auxiliary_loss_clip": 0.01127682,
      "auxiliary_loss_mlp": 0.01025332,
      "balance_loss_clip": 1.04319215,
      "balance_loss_mlp": 1.01785755,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.8565244832430803,
      "language_loss": 0.77853835,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.8000685,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.752708673477173
    },
    {
      "auxiliary_loss_clip": 0.01163364,
      "auxiliary_loss_mlp": 0.00902013,
      "balance_loss_clip": 1.05169618,
      "balance_loss_mlp": 1.00095129,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.8025431014096958,
      "language_loss": 0.76478881,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.78544259,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.590271472930908
    },
    {
      "auxiliary_loss_clip": 0.0115698,
      "auxiliary_loss_mlp": 0.01026505,
      "balance_loss_clip": 1.04551053,
      "balance_loss_mlp": 1.01854813,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 2.092064066846526,
      "language_loss": 0.76760387,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.78943878,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.6379587650299072
    },
    {
      "auxiliary_loss_clip": 0.01172434,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 1.04833412,
      "balance_loss_mlp": 1.01898003,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 1.8775099535599868,
      "language_loss": 0.78002882,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80201864,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.516819715499878
    },
    {
      "auxiliary_loss_clip": 0.01141124,
      "auxiliary_loss_mlp": 0.00901497,
      "balance_loss_clip": 1.04237139,
      "balance_loss_mlp": 1.00081837,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.102580525374874,
      "language_loss": 0.79432136,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81474757,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.6824212074279785
    },
    {
      "auxiliary_loss_clip": 0.01155894,
      "auxiliary_loss_mlp": 0.01025035,
      "balance_loss_clip": 1.04527962,
      "balance_loss_mlp": 1.0182848,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.1366489768867227,
      "language_loss": 0.83208787,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85389721,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.5858733654022217
    },
    {
      "auxiliary_loss_clip": 0.01167379,
      "auxiliary_loss_mlp": 0.01029579,
      "balance_loss_clip": 1.05252326,
      "balance_loss_mlp": 1.02153826,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 2.4677897487493787,
      "language_loss": 0.76414323,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78611284,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 3.5734565258026123
    },
    {
      "auxiliary_loss_clip": 0.01150509,
      "auxiliary_loss_mlp": 0.0102817,
      "balance_loss_clip": 1.04655457,
      "balance_loss_mlp": 1.0203495,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 4.990493435719486,
      "language_loss": 0.65671849,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.6785053,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 2.8888461589813232
    },
    {
      "auxiliary_loss_clip": 0.01159266,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.04658663,
      "balance_loss_mlp": 1.01938462,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 3.9662187228722363,
      "language_loss": 0.7494536,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77131689,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.691157817840576
    },
    {
      "auxiliary_loss_clip": 0.01164295,
      "auxiliary_loss_mlp": 0.01022039,
      "balance_loss_clip": 1.04681158,
      "balance_loss_mlp": 1.01450443,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 4.057888767002114,
      "language_loss": 0.75902367,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.78088701,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.6707236766815186
    },
    {
      "auxiliary_loss_clip": 0.01069996,
      "auxiliary_loss_mlp": 0.01000759,
      "balance_loss_clip": 1.01056123,
      "balance_loss_mlp": 0.99993086,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7759938884032317,
      "language_loss": 0.57460862,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59531623,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.0915613174438477
    },
    {
      "auxiliary_loss_clip": 0.01159255,
      "auxiliary_loss_mlp": 0.01024069,
      "balance_loss_clip": 1.04472995,
      "balance_loss_mlp": 1.01602817,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 6.89295210886588,
      "language_loss": 0.73255187,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75438511,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.5983636379241943
    },
    {
      "auxiliary_loss_clip": 0.01124419,
      "auxiliary_loss_mlp": 0.01030052,
      "balance_loss_clip": 1.04169846,
      "balance_loss_mlp": 1.0225296,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.9214725835116562,
      "language_loss": 0.58079642,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60234106,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.7664954662323
    },
    {
      "auxiliary_loss_clip": 0.01049204,
      "auxiliary_loss_mlp": 0.01002165,
      "balance_loss_clip": 1.01020741,
      "balance_loss_mlp": 1.00130677,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7262287996502907,
      "language_loss": 0.55176729,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57228094,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.3733339309692383
    },
    {
      "auxiliary_loss_clip": 0.0115585,
      "auxiliary_loss_mlp": 0.01023616,
      "balance_loss_clip": 1.04673195,
      "balance_loss_mlp": 1.01547384,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.7370093070346084,
      "language_loss": 0.69749129,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.71928596,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 3.8896045684814453
    },
    {
      "auxiliary_loss_clip": 0.01175502,
      "auxiliary_loss_mlp": 0.01024239,
      "balance_loss_clip": 1.05066073,
      "balance_loss_mlp": 1.01678562,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.8002267726923902,
      "language_loss": 0.71650267,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.73850006,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.6314985752105713
    },
    {
      "auxiliary_loss_clip": 0.01175207,
      "auxiliary_loss_mlp": 0.01028655,
      "balance_loss_clip": 1.05113924,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.1718897894300384,
      "language_loss": 0.68996733,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71200597,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.58846378326416
    },
    {
      "auxiliary_loss_clip": 0.0116143,
      "auxiliary_loss_mlp": 0.01024792,
      "balance_loss_clip": 1.04628682,
      "balance_loss_mlp": 1.01711822,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 2.676700275907767,
      "language_loss": 0.79946494,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.82132721,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.613337516784668
    },
    {
      "auxiliary_loss_clip": 0.01145338,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.04548037,
      "balance_loss_mlp": 1.01954961,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.5980129251226434,
      "language_loss": 0.71896213,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74067867,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.7475690841674805
    },
    {
      "auxiliary_loss_clip": 0.01161401,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.04759717,
      "balance_loss_mlp": 1.01705277,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 3.110626195206033,
      "language_loss": 0.63514888,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65700865,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.560978412628174
    },
    {
      "auxiliary_loss_clip": 0.01163667,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.04659081,
      "balance_loss_mlp": 1.01810861,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.1773228943404055,
      "language_loss": 0.69172359,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71361315,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.6360256671905518
    },
    {
      "auxiliary_loss_clip": 0.01168783,
      "auxiliary_loss_mlp": 0.01024084,
      "balance_loss_clip": 1.04749417,
      "balance_loss_mlp": 1.01661277,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 2.0963184912881756,
      "language_loss": 0.73681879,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75874746,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 3.5144362449645996
    },
    {
      "auxiliary_loss_clip": 0.01151852,
      "auxiliary_loss_mlp": 0.01024047,
      "balance_loss_clip": 1.04611206,
      "balance_loss_mlp": 1.01651895,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 4.443482193635862,
      "language_loss": 0.77802473,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.79978371,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.5922188758850098
    },
    {
      "auxiliary_loss_clip": 0.01163177,
      "auxiliary_loss_mlp": 0.01024804,
      "balance_loss_clip": 1.04812312,
      "balance_loss_mlp": 1.0175705,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 3.716392152844763,
      "language_loss": 0.85977453,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88165438,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 3.577317714691162
    },
    {
      "auxiliary_loss_clip": 0.01163941,
      "auxiliary_loss_mlp": 0.01028048,
      "balance_loss_clip": 1.04706025,
      "balance_loss_mlp": 1.02085948,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.052125595929858,
      "language_loss": 0.76109445,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.78301436,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.559978723526001
    },
    {
      "auxiliary_loss_clip": 0.01057899,
      "auxiliary_loss_mlp": 0.01001952,
      "balance_loss_clip": 1.0108707,
      "balance_loss_mlp": 1.00119162,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7623405008552621,
      "language_loss": 0.6302731,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65087163,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.1511611938476562
    },
    {
      "auxiliary_loss_clip": 0.0113148,
      "auxiliary_loss_mlp": 0.01023332,
      "balance_loss_clip": 1.0418973,
      "balance_loss_mlp": 1.01587284,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.120342009885541,
      "language_loss": 0.81620026,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.83774841,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.6514811515808105
    },
    {
      "auxiliary_loss_clip": 0.01142148,
      "auxiliary_loss_mlp": 0.0102073,
      "balance_loss_clip": 1.04316425,
      "balance_loss_mlp": 1.01387799,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.212560351571794,
      "language_loss": 0.88091224,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.90254104,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.654114246368408
    },
    {
      "auxiliary_loss_clip": 0.01165977,
      "auxiliary_loss_mlp": 0.01021233,
      "balance_loss_clip": 1.04899907,
      "balance_loss_mlp": 1.01392233,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 3.0859016056912965,
      "language_loss": 0.55929172,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58116382,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.6379592418670654
    },
    {
      "auxiliary_loss_clip": 0.01166737,
      "auxiliary_loss_mlp": 0.01022376,
      "balance_loss_clip": 1.04883635,
      "balance_loss_mlp": 1.01544666,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.767650881076986,
      "language_loss": 0.77863234,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.80052346,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.5842185020446777
    },
    {
      "auxiliary_loss_clip": 0.01156813,
      "auxiliary_loss_mlp": 0.01022239,
      "balance_loss_clip": 1.04574025,
      "balance_loss_mlp": 1.01510441,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.5389689473205699,
      "language_loss": 0.80939519,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83118576,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.6676363945007324
    },
    {
      "auxiliary_loss_clip": 0.01155407,
      "auxiliary_loss_mlp": 0.01026837,
      "balance_loss_clip": 1.04985952,
      "balance_loss_mlp": 1.01978564,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.686193368289638,
      "language_loss": 0.69776607,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.71958852,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.6181399822235107
    },
    {
      "auxiliary_loss_clip": 0.01164536,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.04921794,
      "balance_loss_mlp": 1.01821208,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.397718984679882,
      "language_loss": 0.66669643,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68860072,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.708226442337036
    },
    {
      "auxiliary_loss_clip": 0.01164255,
      "auxiliary_loss_mlp": 0.01025024,
      "balance_loss_clip": 1.04677415,
      "balance_loss_mlp": 1.01749575,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 9.497652438629443,
      "language_loss": 0.82674241,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.8486352,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 3.5703258514404297
    },
    {
      "auxiliary_loss_clip": 0.01151464,
      "auxiliary_loss_mlp": 0.0090195,
      "balance_loss_clip": 1.04246819,
      "balance_loss_mlp": 1.00082648,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 2.4842168378951577,
      "language_loss": 0.71073014,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.73126429,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.676891803741455
    },
    {
      "auxiliary_loss_clip": 0.01145522,
      "auxiliary_loss_mlp": 0.0090173,
      "balance_loss_clip": 1.04439902,
      "balance_loss_mlp": 1.00074482,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.2454324267443453,
      "language_loss": 0.70253187,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.72300434,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.7595407962799072
    },
    {
      "auxiliary_loss_clip": 0.01143369,
      "auxiliary_loss_mlp": 0.01022596,
      "balance_loss_clip": 1.04722118,
      "balance_loss_mlp": 1.01511598,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 2.158460231052751,
      "language_loss": 0.69623041,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71789002,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.7542364597320557
    },
    {
      "auxiliary_loss_clip": 0.0115575,
      "auxiliary_loss_mlp": 0.01026076,
      "balance_loss_clip": 1.04560399,
      "balance_loss_mlp": 1.01918888,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.9082390187131397,
      "language_loss": 0.74074364,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76256192,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.6810855865478516
    },
    {
      "auxiliary_loss_clip": 0.01062949,
      "auxiliary_loss_mlp": 0.00890809,
      "balance_loss_clip": 1.01327097,
      "balance_loss_mlp": 1.00004923,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7220557756345044,
      "language_loss": 0.5751906,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59472823,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.202496290206909
    },
    {
      "auxiliary_loss_clip": 0.01150444,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04478002,
      "balance_loss_mlp": 1.01842535,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.392780712018426,
      "language_loss": 0.79211032,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81387693,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.6929266452789307
    },
    {
      "auxiliary_loss_clip": 0.01156431,
      "auxiliary_loss_mlp": 0.01022441,
      "balance_loss_clip": 1.0459404,
      "balance_loss_mlp": 1.0159564,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.6142073024702264,
      "language_loss": 0.77096504,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79275376,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 3.5447347164154053
    },
    {
      "auxiliary_loss_clip": 0.01130457,
      "auxiliary_loss_mlp": 0.0102399,
      "balance_loss_clip": 1.04659534,
      "balance_loss_mlp": 1.0163728,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.9788313086413214,
      "language_loss": 0.88512421,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90666872,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.7665421962738037
    },
    {
      "auxiliary_loss_clip": 0.01165551,
      "auxiliary_loss_mlp": 0.01020712,
      "balance_loss_clip": 1.04763854,
      "balance_loss_mlp": 1.01290131,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 1.8405190049686684,
      "language_loss": 0.72677171,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74863434,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.685086965560913
    },
    {
      "auxiliary_loss_clip": 0.01172171,
      "auxiliary_loss_mlp": 0.01026156,
      "balance_loss_clip": 1.05045116,
      "balance_loss_mlp": 1.01883054,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 4.90738777234658,
      "language_loss": 0.72915757,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.75114083,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.5208754539489746
    },
    {
      "auxiliary_loss_clip": 0.01159791,
      "auxiliary_loss_mlp": 0.0090152,
      "balance_loss_clip": 1.04575646,
      "balance_loss_mlp": 1.00070536,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 2.7435575180289016,
      "language_loss": 0.70463216,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.72524524,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.660861015319824
    },
    {
      "auxiliary_loss_clip": 0.01151814,
      "auxiliary_loss_mlp": 0.01025046,
      "balance_loss_clip": 1.04738855,
      "balance_loss_mlp": 1.01824546,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 2.038834844141726,
      "language_loss": 0.80858219,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.83035076,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.6230850219726562
    },
    {
      "auxiliary_loss_clip": 0.01173007,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.04889166,
      "balance_loss_mlp": 1.01685643,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.170669345965744,
      "language_loss": 0.67320716,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69518602,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.5680794715881348
    },
    {
      "auxiliary_loss_clip": 0.01127865,
      "auxiliary_loss_mlp": 0.01025534,
      "balance_loss_clip": 1.03891194,
      "balance_loss_mlp": 1.01824749,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 2.1676421344355865,
      "language_loss": 0.71325731,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.73479128,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 3.580486297607422
    },
    {
      "auxiliary_loss_clip": 0.0114456,
      "auxiliary_loss_mlp": 0.01021809,
      "balance_loss_clip": 1.04531491,
      "balance_loss_mlp": 1.014364,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 3.245786305110208,
      "language_loss": 0.74340349,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76506716,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.7239537239074707
    },
    {
      "auxiliary_loss_clip": 0.01149055,
      "auxiliary_loss_mlp": 0.01031485,
      "balance_loss_clip": 1.04574609,
      "balance_loss_mlp": 1.02387965,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 2.7466528985222687,
      "language_loss": 0.81701016,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83881557,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 3.673234224319458
    },
    {
      "auxiliary_loss_clip": 0.01140941,
      "auxiliary_loss_mlp": 0.01021912,
      "balance_loss_clip": 1.04545975,
      "balance_loss_mlp": 1.01509309,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.8750585903763852,
      "language_loss": 0.76485187,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78648043,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.6962852478027344
    },
    {
      "auxiliary_loss_clip": 0.01159423,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04716003,
      "balance_loss_mlp": 1.01921654,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 4.101928614245097,
      "language_loss": 0.74465281,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76650512,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.750849723815918
    },
    {
      "auxiliary_loss_clip": 0.01159961,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 1.04726851,
      "balance_loss_mlp": 1.02152228,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.4640768551048255,
      "language_loss": 0.72168726,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74357378,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.594193696975708
    },
    {
      "auxiliary_loss_clip": 0.01146712,
      "auxiliary_loss_mlp": 0.00900605,
      "balance_loss_clip": 1.04307115,
      "balance_loss_mlp": 1.00067067,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 1.8945134306913476,
      "language_loss": 0.8005234,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.82099664,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.6929004192352295
    },
    {
      "auxiliary_loss_clip": 0.01169287,
      "auxiliary_loss_mlp": 0.01021861,
      "balance_loss_clip": 1.04880905,
      "balance_loss_mlp": 1.01414561,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.7168402336945374,
      "language_loss": 0.68182957,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70374107,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.548473596572876
    },
    {
      "auxiliary_loss_clip": 0.01173573,
      "auxiliary_loss_mlp": 0.01025385,
      "balance_loss_clip": 1.0496192,
      "balance_loss_mlp": 1.01786852,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 2.1070122953205095,
      "language_loss": 0.77003431,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.7920239,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.5737037658691406
    },
    {
      "auxiliary_loss_clip": 0.01141874,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04398501,
      "balance_loss_mlp": 1.0157249,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 2.186168948262134,
      "language_loss": 0.70175922,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72341043,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.741384267807007
    },
    {
      "auxiliary_loss_clip": 0.01160076,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.05119169,
      "balance_loss_mlp": 1.01800096,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.5399550195308698,
      "language_loss": 0.82121664,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.84307271,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.6753499507904053
    },
    {
      "auxiliary_loss_clip": 0.01169765,
      "auxiliary_loss_mlp": 0.00901584,
      "balance_loss_clip": 1.04931235,
      "balance_loss_mlp": 1.00069332,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.7686868453181683,
      "language_loss": 0.79058492,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.81129843,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 3.533013343811035
    },
    {
      "auxiliary_loss_clip": 0.01127732,
      "auxiliary_loss_mlp": 0.0102221,
      "balance_loss_clip": 1.04132318,
      "balance_loss_mlp": 1.01556706,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.085108725860661,
      "language_loss": 0.70944136,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.73094076,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.941815137863159
    },
    {
      "auxiliary_loss_clip": 0.01166901,
      "auxiliary_loss_mlp": 0.01024655,
      "balance_loss_clip": 1.04602826,
      "balance_loss_mlp": 1.01705587,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 1.8362433941522622,
      "language_loss": 0.62883288,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.65074849,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.7925233840942383
    },
    {
      "auxiliary_loss_clip": 0.01143954,
      "auxiliary_loss_mlp": 0.01023031,
      "balance_loss_clip": 1.0413847,
      "balance_loss_mlp": 1.01552105,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.2708687638878433,
      "language_loss": 0.76935965,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.79102951,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.639625310897827
    },
    {
      "auxiliary_loss_clip": 0.01163424,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.04456413,
      "balance_loss_mlp": 1.01705265,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 1.9087920046819458,
      "language_loss": 0.78823549,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.8101145,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.616689920425415
    },
    {
      "auxiliary_loss_clip": 0.01140112,
      "auxiliary_loss_mlp": 0.01021432,
      "balance_loss_clip": 1.04201555,
      "balance_loss_mlp": 1.01412702,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 2.1080766824630284,
      "language_loss": 0.77884823,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80046368,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.7317912578582764
    },
    {
      "auxiliary_loss_clip": 0.01152001,
      "auxiliary_loss_mlp": 0.00901454,
      "balance_loss_clip": 1.04348266,
      "balance_loss_mlp": 1.00083947,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.192376967140394,
      "language_loss": 0.74556243,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.76609701,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.6796960830688477
    },
    {
      "auxiliary_loss_clip": 0.0116224,
      "auxiliary_loss_mlp": 0.01028503,
      "balance_loss_clip": 1.04731834,
      "balance_loss_mlp": 1.02179694,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 3.0351631953333924,
      "language_loss": 0.73877025,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76067764,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.6158711910247803
    },
    {
      "auxiliary_loss_clip": 0.01138081,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.04550385,
      "balance_loss_mlp": 1.02221632,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 2.055560202375853,
      "language_loss": 0.64548624,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66715574,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.653855323791504
    },
    {
      "auxiliary_loss_clip": 0.01146075,
      "auxiliary_loss_mlp": 0.01021603,
      "balance_loss_clip": 1.04585981,
      "balance_loss_mlp": 1.01425338,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 2.1756039726479894,
      "language_loss": 0.88470685,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90638363,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 3.5327870845794678
    },
    {
      "auxiliary_loss_clip": 0.01161789,
      "auxiliary_loss_mlp": 0.01023168,
      "balance_loss_clip": 1.04973996,
      "balance_loss_mlp": 1.01578557,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 2.8769593990725637,
      "language_loss": 0.66655385,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.68840349,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.6582839488983154
    },
    {
      "auxiliary_loss_clip": 0.01165655,
      "auxiliary_loss_mlp": 0.0090129,
      "balance_loss_clip": 1.04817152,
      "balance_loss_mlp": 1.00072312,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.7096034403380773,
      "language_loss": 0.81001949,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83068895,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.683368682861328
    },
    {
      "auxiliary_loss_clip": 0.01159327,
      "auxiliary_loss_mlp": 0.01021427,
      "balance_loss_clip": 1.04514265,
      "balance_loss_mlp": 1.01433468,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 2.365546091265705,
      "language_loss": 0.80115426,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82296181,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.605574607849121
    },
    {
      "auxiliary_loss_clip": 0.01150038,
      "auxiliary_loss_mlp": 0.01021681,
      "balance_loss_clip": 1.04689074,
      "balance_loss_mlp": 1.01462042,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.8963228999290143,
      "language_loss": 0.70300531,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.7247225,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.6300787925720215
    },
    {
      "auxiliary_loss_clip": 0.0107026,
      "auxiliary_loss_mlp": 0.01000669,
      "balance_loss_clip": 1.01081467,
      "balance_loss_mlp": 0.99988192,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8053307878317539,
      "language_loss": 0.60562938,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62633866,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.186035633087158
    },
    {
      "auxiliary_loss_clip": 0.01158646,
      "auxiliary_loss_mlp": 0.01025374,
      "balance_loss_clip": 1.04636574,
      "balance_loss_mlp": 1.01733387,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.9738815486610994,
      "language_loss": 0.71967947,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.74151969,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.6708099842071533
    },
    {
      "auxiliary_loss_clip": 0.011752,
      "auxiliary_loss_mlp": 0.01025008,
      "balance_loss_clip": 1.05293417,
      "balance_loss_mlp": 1.01838279,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.828549433749723,
      "language_loss": 0.62649781,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64849991,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 3.549983501434326
    },
    {
      "auxiliary_loss_clip": 0.01135357,
      "auxiliary_loss_mlp": 0.01024511,
      "balance_loss_clip": 1.04537535,
      "balance_loss_mlp": 1.01719213,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 2.0230155584991647,
      "language_loss": 0.85258973,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87418836,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.704902410507202
    },
    {
      "auxiliary_loss_clip": 0.01162039,
      "auxiliary_loss_mlp": 0.01028931,
      "balance_loss_clip": 1.0459578,
      "balance_loss_mlp": 1.02162027,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 2.0244603846327216,
      "language_loss": 0.68343681,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70534652,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 3.568532943725586
    },
    {
      "auxiliary_loss_clip": 0.0115217,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.04849792,
      "balance_loss_mlp": 1.01474428,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.1235409973304025,
      "language_loss": 0.81117374,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83291936,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.623608350753784
    },
    {
      "auxiliary_loss_clip": 0.01153731,
      "auxiliary_loss_mlp": 0.01027453,
      "balance_loss_clip": 1.04495072,
      "balance_loss_mlp": 1.02001441,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.240439625183069,
      "language_loss": 0.75147355,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77328539,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.6125307083129883
    },
    {
      "auxiliary_loss_clip": 0.01149856,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.04324055,
      "balance_loss_mlp": 1.01630998,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.2907449403676887,
      "language_loss": 0.7030338,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72477108,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.578742265701294
    },
    {
      "auxiliary_loss_clip": 0.01150375,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.04683757,
      "balance_loss_mlp": 1.02458572,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.5811236154980528,
      "language_loss": 0.73352945,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75535095,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.7582321166992188
    },
    {
      "auxiliary_loss_clip": 0.01170966,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.05054832,
      "balance_loss_mlp": 1.02073598,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.404717310809306,
      "language_loss": 0.83498859,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85697532,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.5139098167419434
    },
    {
      "auxiliary_loss_clip": 0.01158769,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.04582083,
      "balance_loss_mlp": 1.02112913,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 1.8948940507282883,
      "language_loss": 0.66506451,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68693399,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.6563987731933594
    },
    {
      "auxiliary_loss_clip": 0.01175014,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.04910851,
      "balance_loss_mlp": 1.01952922,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.4144305885996165,
      "language_loss": 0.79085374,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81288075,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.5669822692871094
    },
    {
      "auxiliary_loss_clip": 0.01143779,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04461372,
      "balance_loss_mlp": 1.01967359,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 2.0786059562679777,
      "language_loss": 0.74634498,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76805329,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.6659998893737793
    },
    {
      "auxiliary_loss_clip": 0.01159149,
      "auxiliary_loss_mlp": 0.01026677,
      "balance_loss_clip": 1.04413629,
      "balance_loss_mlp": 1.01861882,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.8712311163909696,
      "language_loss": 0.71529531,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73715353,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 3.521054267883301
    },
    {
      "auxiliary_loss_clip": 0.01068378,
      "auxiliary_loss_mlp": 0.01002607,
      "balance_loss_clip": 1.02540469,
      "balance_loss_mlp": 1.00168276,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9675966608951714,
      "language_loss": 0.61175722,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63246709,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.722904682159424
    },
    {
      "auxiliary_loss_clip": 0.01141377,
      "auxiliary_loss_mlp": 0.00901214,
      "balance_loss_clip": 1.04711604,
      "balance_loss_mlp": 1.00074458,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.8684175543883663,
      "language_loss": 0.72869307,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.74911898,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 3.029947519302368
    },
    {
      "auxiliary_loss_clip": 0.01144579,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.04505599,
      "balance_loss_mlp": 1.01788127,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 2.0524365479803093,
      "language_loss": 0.70233178,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72402883,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.6805903911590576
    },
    {
      "auxiliary_loss_clip": 0.01170675,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.04962385,
      "balance_loss_mlp": 1.01770186,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 4.150948734252798,
      "language_loss": 0.77891064,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.80086327,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.5414912700653076
    },
    {
      "auxiliary_loss_clip": 0.01170033,
      "auxiliary_loss_mlp": 0.01023682,
      "balance_loss_clip": 1.04792941,
      "balance_loss_mlp": 1.01640463,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.144934244429937,
      "language_loss": 0.84554315,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86748034,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.582138776779175
    },
    {
      "auxiliary_loss_clip": 0.0111908,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.03986263,
      "balance_loss_mlp": 1.01790273,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 2.646807192744607,
      "language_loss": 0.71195292,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.7333951,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.7765116691589355
    },
    {
      "auxiliary_loss_clip": 0.01168361,
      "auxiliary_loss_mlp": 0.01027337,
      "balance_loss_clip": 1.04893756,
      "balance_loss_mlp": 1.01998186,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 2.383595966417527,
      "language_loss": 0.65508473,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67704171,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.649273633956909
    },
    {
      "auxiliary_loss_clip": 0.0114186,
      "auxiliary_loss_mlp": 0.01025476,
      "balance_loss_clip": 1.04404163,
      "balance_loss_mlp": 1.01812387,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 1.986239921538006,
      "language_loss": 0.82294941,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84462279,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 3.7095327377319336
    },
    {
      "auxiliary_loss_clip": 0.01151394,
      "auxiliary_loss_mlp": 0.01025775,
      "balance_loss_clip": 1.04636526,
      "balance_loss_mlp": 1.01849747,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.830842042364391,
      "language_loss": 0.79616737,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81793904,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.7138781547546387
    },
    {
      "auxiliary_loss_clip": 0.01055083,
      "auxiliary_loss_mlp": 0.00891138,
      "balance_loss_clip": 1.0108819,
      "balance_loss_mlp": 0.99989623,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7966453135406537,
      "language_loss": 0.56491995,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58438212,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.237614393234253
    },
    {
      "auxiliary_loss_clip": 0.01164993,
      "auxiliary_loss_mlp": 0.01028054,
      "balance_loss_clip": 1.05333185,
      "balance_loss_mlp": 1.02054107,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.6333745598155602,
      "language_loss": 0.76533967,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78727013,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.6178150177001953
    },
    {
      "auxiliary_loss_clip": 0.0117208,
      "auxiliary_loss_mlp": 0.0102222,
      "balance_loss_clip": 1.05073559,
      "balance_loss_mlp": 1.01429868,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 2.002993919630587,
      "language_loss": 0.85533941,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87728238,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 2.749467372894287
    },
    {
      "auxiliary_loss_clip": 0.01165145,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.04702187,
      "balance_loss_mlp": 1.01526749,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.8194907094014499,
      "language_loss": 0.64840662,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.67028546,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.6609933376312256
    },
    {
      "auxiliary_loss_clip": 0.01173036,
      "auxiliary_loss_mlp": 0.01024795,
      "balance_loss_clip": 1.05075812,
      "balance_loss_mlp": 1.01717806,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.277932289530777,
      "language_loss": 0.73249424,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75447261,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.547776222229004
    },
    {
      "auxiliary_loss_clip": 0.01150353,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 1.04531765,
      "balance_loss_mlp": 1.01794124,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.7308838102171973,
      "language_loss": 0.77434111,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79609579,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 3.614426612854004
    },
    {
      "auxiliary_loss_clip": 0.01163989,
      "auxiliary_loss_mlp": 0.01024225,
      "balance_loss_clip": 1.04785037,
      "balance_loss_mlp": 1.01693511,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.385217181532153,
      "language_loss": 0.83728987,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85917199,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.5574069023132324
    },
    {
      "auxiliary_loss_clip": 0.01156851,
      "auxiliary_loss_mlp": 0.01028482,
      "balance_loss_clip": 1.04506004,
      "balance_loss_mlp": 1.02085853,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 2.1047826388184765,
      "language_loss": 0.64118755,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66304088,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.555835723876953
    },
    {
      "auxiliary_loss_clip": 0.01143754,
      "auxiliary_loss_mlp": 0.01019854,
      "balance_loss_clip": 1.04312503,
      "balance_loss_mlp": 1.01281166,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.9008084111854218,
      "language_loss": 0.72771835,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74935442,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 3.624512195587158
    },
    {
      "auxiliary_loss_clip": 0.01165826,
      "auxiliary_loss_mlp": 0.01024471,
      "balance_loss_clip": 1.04782426,
      "balance_loss_mlp": 1.01697016,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.6121673716103246,
      "language_loss": 0.75755644,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77945942,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.5604634284973145
    },
    {
      "auxiliary_loss_clip": 0.01160162,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.04577088,
      "balance_loss_mlp": 1.0196619,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 2.00443920545741,
      "language_loss": 0.65906173,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.68093479,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.74794864654541
    },
    {
      "auxiliary_loss_clip": 0.01137336,
      "auxiliary_loss_mlp": 0.01025887,
      "balance_loss_clip": 1.04696798,
      "balance_loss_mlp": 1.01830518,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 7.504384654138385,
      "language_loss": 0.79071921,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.81235147,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.745750904083252
    },
    {
      "auxiliary_loss_clip": 0.01158841,
      "auxiliary_loss_mlp": 0.01024716,
      "balance_loss_clip": 1.04721522,
      "balance_loss_mlp": 1.01767981,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.770881591642932,
      "language_loss": 0.77310765,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79494321,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.64631986618042
    },
    {
      "auxiliary_loss_clip": 0.01168555,
      "auxiliary_loss_mlp": 0.01022104,
      "balance_loss_clip": 1.0493474,
      "balance_loss_mlp": 1.01485646,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 3.029585983953228,
      "language_loss": 0.91307002,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.9349767,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.6628360748291016
    },
    {
      "auxiliary_loss_clip": 0.01158535,
      "auxiliary_loss_mlp": 0.01026026,
      "balance_loss_clip": 1.04608798,
      "balance_loss_mlp": 1.01940107,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.5676462567135232,
      "language_loss": 0.75034374,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77218926,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.5738942623138428
    },
    {
      "auxiliary_loss_clip": 0.01136519,
      "auxiliary_loss_mlp": 0.01022585,
      "balance_loss_clip": 1.04280782,
      "balance_loss_mlp": 1.01536131,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 2.242052135006912,
      "language_loss": 0.82551301,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.84710407,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.7145395278930664
    },
    {
      "auxiliary_loss_clip": 0.01160515,
      "auxiliary_loss_mlp": 0.01021685,
      "balance_loss_clip": 1.0470736,
      "balance_loss_mlp": 1.01453233,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.8599667889022615,
      "language_loss": 0.75873768,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78055966,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 3.6682937145233154
    },
    {
      "auxiliary_loss_clip": 0.01138006,
      "auxiliary_loss_mlp": 0.01023026,
      "balance_loss_clip": 1.04218102,
      "balance_loss_mlp": 1.01578403,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 2.011631180518327,
      "language_loss": 0.75104141,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.77265167,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.6479923725128174
    },
    {
      "auxiliary_loss_clip": 0.0113945,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.04409862,
      "balance_loss_mlp": 1.01647568,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.2352893789874715,
      "language_loss": 0.84296083,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.8645975,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.8364479541778564
    },
    {
      "auxiliary_loss_clip": 0.01149423,
      "auxiliary_loss_mlp": 0.01021609,
      "balance_loss_clip": 1.04786825,
      "balance_loss_mlp": 1.01502013,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.8825599499155317,
      "language_loss": 0.80309093,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82480127,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.661052703857422
    },
    {
      "auxiliary_loss_clip": 0.01149739,
      "auxiliary_loss_mlp": 0.01025979,
      "balance_loss_clip": 1.04360306,
      "balance_loss_mlp": 1.01837325,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 2.1470514284462725,
      "language_loss": 0.86104155,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.88279873,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.628354787826538
    },
    {
      "auxiliary_loss_clip": 0.01171845,
      "auxiliary_loss_mlp": 0.01026382,
      "balance_loss_clip": 1.04895329,
      "balance_loss_mlp": 1.01880312,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 2.0970255601202528,
      "language_loss": 0.63445783,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65644008,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.6027567386627197
    },
    {
      "auxiliary_loss_clip": 0.01139896,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 1.04418004,
      "balance_loss_mlp": 1.01943254,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.767806730426881,
      "language_loss": 0.69251949,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71418321,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.6901376247406006
    },
    {
      "auxiliary_loss_clip": 0.01149981,
      "auxiliary_loss_mlp": 0.01022482,
      "balance_loss_clip": 1.04284644,
      "balance_loss_mlp": 1.01535368,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 2.0503466679654583,
      "language_loss": 0.74662054,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76834518,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.614522695541382
    },
    {
      "auxiliary_loss_clip": 0.0116439,
      "auxiliary_loss_mlp": 0.01030295,
      "balance_loss_clip": 1.04769564,
      "balance_loss_mlp": 1.02248645,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.9318869783215835,
      "language_loss": 0.82318246,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84512931,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.6605758666992188
    },
    {
      "auxiliary_loss_clip": 0.01163313,
      "auxiliary_loss_mlp": 0.01029478,
      "balance_loss_clip": 1.04912245,
      "balance_loss_mlp": 1.02226019,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.829946007077107,
      "language_loss": 0.76545763,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78738558,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 3.5616886615753174
    },
    {
      "auxiliary_loss_clip": 0.01168549,
      "auxiliary_loss_mlp": 0.01023089,
      "balance_loss_clip": 1.04831624,
      "balance_loss_mlp": 1.01608276,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 2.074104094037146,
      "language_loss": 0.85924435,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.88116074,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.59501576423645
    },
    {
      "auxiliary_loss_clip": 0.0116318,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.04940271,
      "balance_loss_mlp": 1.01902354,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.5731133148852867,
      "language_loss": 0.79117036,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81306052,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.716567277908325
    },
    {
      "auxiliary_loss_clip": 0.01172957,
      "auxiliary_loss_mlp": 0.01023848,
      "balance_loss_clip": 1.05036974,
      "balance_loss_mlp": 1.01620626,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 2.278669516487258,
      "language_loss": 0.85438401,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87635207,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 2.5358774662017822
    },
    {
      "auxiliary_loss_clip": 0.01144314,
      "auxiliary_loss_mlp": 0.01027766,
      "balance_loss_clip": 1.04391003,
      "balance_loss_mlp": 1.02095294,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.2399950590787463,
      "language_loss": 0.69212568,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.7138465,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.713843822479248
    },
    {
      "auxiliary_loss_clip": 0.0114807,
      "auxiliary_loss_mlp": 0.01024639,
      "balance_loss_clip": 1.04266274,
      "balance_loss_mlp": 1.01695037,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.097892099833657,
      "language_loss": 0.79225028,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81397736,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.6234004497528076
    },
    {
      "auxiliary_loss_clip": 0.01153076,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.04359257,
      "balance_loss_mlp": 1.02051854,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.7724715215336972,
      "language_loss": 0.84810996,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.86991847,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.648021936416626
    },
    {
      "auxiliary_loss_clip": 0.01154756,
      "auxiliary_loss_mlp": 0.01027629,
      "balance_loss_clip": 1.04470968,
      "balance_loss_mlp": 1.02008271,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 2.161528511487358,
      "language_loss": 0.68918395,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71100777,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 3.585331678390503
    },
    {
      "auxiliary_loss_clip": 0.01159197,
      "auxiliary_loss_mlp": 0.0102368,
      "balance_loss_clip": 1.04792929,
      "balance_loss_mlp": 1.01630974,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 2.705475251411954,
      "language_loss": 0.7479974,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.76982617,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.6189358234405518
    },
    {
      "auxiliary_loss_clip": 0.011729,
      "auxiliary_loss_mlp": 0.01034574,
      "balance_loss_clip": 1.0497067,
      "balance_loss_mlp": 1.02720952,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.197563476390904,
      "language_loss": 0.69033521,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.71240997,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.659482479095459
    },
    {
      "auxiliary_loss_clip": 0.01149657,
      "auxiliary_loss_mlp": 0.01020871,
      "balance_loss_clip": 1.04485095,
      "balance_loss_mlp": 1.01368821,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.657289795874522,
      "language_loss": 0.75396729,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77567255,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 3.564643383026123
    },
    {
      "auxiliary_loss_clip": 0.01158413,
      "auxiliary_loss_mlp": 0.00901237,
      "balance_loss_clip": 1.04944515,
      "balance_loss_mlp": 1.0007689,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 7.767259080304315,
      "language_loss": 0.79378027,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81437677,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.6299939155578613
    },
    {
      "auxiliary_loss_clip": 0.01162139,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.04858923,
      "balance_loss_mlp": 1.02144027,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.589896424848582,
      "language_loss": 0.66057974,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68249846,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.568803548812866
    },
    {
      "auxiliary_loss_clip": 0.01139688,
      "auxiliary_loss_mlp": 0.01021951,
      "balance_loss_clip": 1.04465604,
      "balance_loss_mlp": 1.0152272,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.8209067546766657,
      "language_loss": 0.79429829,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81591469,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.6730422973632812
    },
    {
      "auxiliary_loss_clip": 0.01144629,
      "auxiliary_loss_mlp": 0.01024753,
      "balance_loss_clip": 1.04132664,
      "balance_loss_mlp": 1.01778793,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 2.0954092043989174,
      "language_loss": 0.67557329,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69726712,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.7052931785583496
    },
    {
      "auxiliary_loss_clip": 0.01056806,
      "auxiliary_loss_mlp": 0.01001082,
      "balance_loss_clip": 1.00925624,
      "balance_loss_mlp": 1.00026572,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7175742048799899,
      "language_loss": 0.58584237,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60642135,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.216247320175171
    },
    {
      "auxiliary_loss_clip": 0.01148483,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.04693913,
      "balance_loss_mlp": 1.01855803,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 2.335359830357137,
      "language_loss": 0.72003329,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.74178076,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.674879550933838
    },
    {
      "auxiliary_loss_clip": 0.01164045,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.04738212,
      "balance_loss_mlp": 1.01688802,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 1.676743817001572,
      "language_loss": 0.75816721,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78005123,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.5995328426361084
    },
    {
      "auxiliary_loss_clip": 0.0115344,
      "auxiliary_loss_mlp": 0.01032316,
      "balance_loss_clip": 1.04518056,
      "balance_loss_mlp": 1.02523255,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 1.9538208697330885,
      "language_loss": 0.80394936,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82580698,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 3.6664185523986816
    },
    {
      "auxiliary_loss_clip": 0.01173757,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.04922795,
      "balance_loss_mlp": 1.02398658,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 2.7924219628603524,
      "language_loss": 0.69304395,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71510011,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.509885549545288
    },
    {
      "auxiliary_loss_clip": 0.0114619,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.04511142,
      "balance_loss_mlp": 1.0194943,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.9154089718930192,
      "language_loss": 0.78515166,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80687833,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.7956268787384033
    },
    {
      "auxiliary_loss_clip": 0.01139396,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.04395819,
      "balance_loss_mlp": 1.01842821,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.7066185114415175,
      "language_loss": 0.73501527,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75666642,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.7952685356140137
    },
    {
      "auxiliary_loss_clip": 0.01152827,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04632497,
      "balance_loss_mlp": 1.02017903,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.756124722414965,
      "language_loss": 0.78722155,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80902648,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.5882115364074707
    },
    {
      "auxiliary_loss_clip": 0.01153429,
      "auxiliary_loss_mlp": 0.01021145,
      "balance_loss_clip": 1.04906416,
      "balance_loss_mlp": 1.01352155,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.368600197030076,
      "language_loss": 0.7871021,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80884784,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.690342664718628
    },
    {
      "auxiliary_loss_clip": 0.01147521,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.04749978,
      "balance_loss_mlp": 1.01884937,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.8235375049249565,
      "language_loss": 0.83894169,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.86067176,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.6260569095611572
    },
    {
      "auxiliary_loss_clip": 0.01154592,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.04651237,
      "balance_loss_mlp": 1.01980639,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 2.5763637551238725,
      "language_loss": 0.75664818,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.77846575,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.6758861541748047
    },
    {
      "auxiliary_loss_clip": 0.01149933,
      "auxiliary_loss_mlp": 0.01022178,
      "balance_loss_clip": 1.04201567,
      "balance_loss_mlp": 1.01473379,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 3.6937822708645953,
      "language_loss": 0.56195188,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.583673,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 3.4866883754730225
    },
    {
      "auxiliary_loss_clip": 0.01159106,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 1.04624367,
      "balance_loss_mlp": 1.01846671,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.939398006422067,
      "language_loss": 0.76831937,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79017133,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.6583213806152344
    },
    {
      "auxiliary_loss_clip": 0.01119295,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.04004359,
      "balance_loss_mlp": 1.02418804,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.407293122181709,
      "language_loss": 0.8155638,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83707964,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.7207818031311035
    },
    {
      "auxiliary_loss_clip": 0.01147795,
      "auxiliary_loss_mlp": 0.01023332,
      "balance_loss_clip": 1.04215527,
      "balance_loss_mlp": 1.01595259,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.36497381375489,
      "language_loss": 0.70637769,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.72808897,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.651503086090088
    },
    {
      "auxiliary_loss_clip": 0.0112478,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.03806388,
      "balance_loss_mlp": 1.01810896,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.4413700557218183,
      "language_loss": 0.74361098,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.76510948,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.7163050174713135
    },
    {
      "auxiliary_loss_clip": 0.01165497,
      "auxiliary_loss_mlp": 0.01024493,
      "balance_loss_clip": 1.04707456,
      "balance_loss_mlp": 1.01656604,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 7.4740881380077075,
      "language_loss": 0.76456565,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.78646559,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.6550474166870117
    },
    {
      "auxiliary_loss_clip": 0.01166061,
      "auxiliary_loss_mlp": 0.01026066,
      "balance_loss_clip": 1.04521012,
      "balance_loss_mlp": 1.01836562,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.378562107383133,
      "language_loss": 0.69381118,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71573246,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.623561143875122
    },
    {
      "auxiliary_loss_clip": 0.01166689,
      "auxiliary_loss_mlp": 0.01021418,
      "balance_loss_clip": 1.04775095,
      "balance_loss_mlp": 1.01409817,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 2.483240896024101,
      "language_loss": 0.7623024,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.7841835,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.621471405029297
    },
    {
      "auxiliary_loss_clip": 0.01149511,
      "auxiliary_loss_mlp": 0.01026008,
      "balance_loss_clip": 1.05027783,
      "balance_loss_mlp": 1.01893926,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.7860052422765034,
      "language_loss": 0.68437886,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70613408,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 3.5837392807006836
    },
    {
      "auxiliary_loss_clip": 0.01051084,
      "auxiliary_loss_mlp": 0.0100219,
      "balance_loss_clip": 1.010391,
      "balance_loss_mlp": 1.00140321,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7773502879800519,
      "language_loss": 0.5447287,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56526148,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.4743967056274414
    },
    {
      "auxiliary_loss_clip": 0.01161917,
      "auxiliary_loss_mlp": 0.01024411,
      "balance_loss_clip": 1.04894555,
      "balance_loss_mlp": 1.01783693,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.4191241700860644,
      "language_loss": 0.74434173,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76620501,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 3.715104818344116
    },
    {
      "auxiliary_loss_clip": 0.01160421,
      "auxiliary_loss_mlp": 0.01030205,
      "balance_loss_clip": 1.04842257,
      "balance_loss_mlp": 1.02342463,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 3.059283787519963,
      "language_loss": 0.79458171,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81648797,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.7805581092834473
    },
    {
      "auxiliary_loss_clip": 0.0106411,
      "auxiliary_loss_mlp": 0.01001765,
      "balance_loss_clip": 1.01281881,
      "balance_loss_mlp": 1.00093102,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7185256614822663,
      "language_loss": 0.56662512,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58728385,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 3.0129995346069336
    },
    {
      "auxiliary_loss_clip": 0.01163428,
      "auxiliary_loss_mlp": 0.01023516,
      "balance_loss_clip": 1.05038118,
      "balance_loss_mlp": 1.01633716,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 2.236732420032336,
      "language_loss": 0.76355201,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78542149,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.669044017791748
    },
    {
      "auxiliary_loss_clip": 0.01149517,
      "auxiliary_loss_mlp": 0.01023704,
      "balance_loss_clip": 1.04648876,
      "balance_loss_mlp": 1.01623559,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.7918758437299043,
      "language_loss": 0.71001995,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.73175216,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.6555838584899902
    },
    {
      "auxiliary_loss_clip": 0.01149943,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.04325771,
      "balance_loss_mlp": 1.02085149,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 2.033804876869811,
      "language_loss": 0.81369501,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.8354786,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.643155574798584
    },
    {
      "auxiliary_loss_clip": 0.01170767,
      "auxiliary_loss_mlp": 0.01023289,
      "balance_loss_clip": 1.04704571,
      "balance_loss_mlp": 1.01562142,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 2.2418321449979834,
      "language_loss": 0.79185784,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.81379837,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.5805985927581787
    },
    {
      "auxiliary_loss_clip": 0.01151902,
      "auxiliary_loss_mlp": 0.01022094,
      "balance_loss_clip": 1.04422426,
      "balance_loss_mlp": 1.01460791,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 2.354191717924115,
      "language_loss": 0.7537781,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77551806,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.5632433891296387
    },
    {
      "auxiliary_loss_clip": 0.01141714,
      "auxiliary_loss_mlp": 0.01025096,
      "balance_loss_clip": 1.04618204,
      "balance_loss_mlp": 1.01814318,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 1.9435517225946208,
      "language_loss": 0.76996255,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.79163074,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 3.6022098064422607
    },
    {
      "auxiliary_loss_clip": 0.011428,
      "auxiliary_loss_mlp": 0.01029147,
      "balance_loss_clip": 1.04358482,
      "balance_loss_mlp": 1.02200341,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 1.942561417138284,
      "language_loss": 0.68337107,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70509052,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.7822744846343994
    },
    {
      "auxiliary_loss_clip": 0.011491,
      "auxiliary_loss_mlp": 0.01022822,
      "balance_loss_clip": 1.04255986,
      "balance_loss_mlp": 1.01491213,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 2.167582983802003,
      "language_loss": 0.75758433,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77930355,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.670309543609619
    },
    {
      "auxiliary_loss_clip": 0.01149747,
      "auxiliary_loss_mlp": 0.01026143,
      "balance_loss_clip": 1.04881859,
      "balance_loss_mlp": 1.0187459,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.711572986099246,
      "language_loss": 0.80084944,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.82260835,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.677600383758545
    },
    {
      "auxiliary_loss_clip": 0.01169841,
      "auxiliary_loss_mlp": 0.01023409,
      "balance_loss_clip": 1.04936111,
      "balance_loss_mlp": 1.01624417,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.7441814953120947,
      "language_loss": 0.79909277,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.82102525,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.5584983825683594
    },
    {
      "auxiliary_loss_clip": 0.01116426,
      "auxiliary_loss_mlp": 0.01021602,
      "balance_loss_clip": 1.03676295,
      "balance_loss_mlp": 1.01440811,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.695354916136069,
      "language_loss": 0.61859649,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63997674,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.8434340953826904
    },
    {
      "auxiliary_loss_clip": 0.01112546,
      "auxiliary_loss_mlp": 0.01022349,
      "balance_loss_clip": 1.04125047,
      "balance_loss_mlp": 1.01578689,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.9209904770760469,
      "language_loss": 0.69351697,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71486592,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.7797508239746094
    },
    {
      "auxiliary_loss_clip": 0.01158173,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.04691958,
      "balance_loss_mlp": 1.01876485,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.286290477925503,
      "language_loss": 0.71862257,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74046981,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.652977466583252
    },
    {
      "auxiliary_loss_clip": 0.0105262,
      "auxiliary_loss_mlp": 0.01001185,
      "balance_loss_clip": 1.0125879,
      "balance_loss_mlp": 1.00040388,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.7997897350064449,
      "language_loss": 0.58036011,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60089815,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 4.179959297180176
    },
    {
      "auxiliary_loss_clip": 0.01131225,
      "auxiliary_loss_mlp": 0.00900972,
      "balance_loss_clip": 1.04324174,
      "balance_loss_mlp": 1.0008316,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 2.0569643933149795,
      "language_loss": 0.60302377,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.62334573,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.752711296081543
    },
    {
      "auxiliary_loss_clip": 0.01144437,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.04303741,
      "balance_loss_mlp": 1.01627994,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.190790813085071,
      "language_loss": 0.67227745,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.69395244,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.5972025394439697
    },
    {
      "auxiliary_loss_clip": 0.01154813,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.04755092,
      "balance_loss_mlp": 1.01968813,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 2.469488233370298,
      "language_loss": 0.68463457,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.70645726,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.8238842487335205
    },
    {
      "auxiliary_loss_clip": 0.01144393,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.04603767,
      "balance_loss_mlp": 1.02056074,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 1.6963619538391967,
      "language_loss": 0.82594943,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.8476671,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.694633722305298
    },
    {
      "auxiliary_loss_clip": 0.01160646,
      "auxiliary_loss_mlp": 0.00900717,
      "balance_loss_clip": 1.04860568,
      "balance_loss_mlp": 1.00076103,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 1.6841716524038313,
      "language_loss": 0.75938654,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.78000021,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.5635037422180176
    },
    {
      "auxiliary_loss_clip": 0.01130139,
      "auxiliary_loss_mlp": 0.0102184,
      "balance_loss_clip": 1.03939831,
      "balance_loss_mlp": 1.01486611,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 2.1642116594533913,
      "language_loss": 0.77568728,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79720706,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.6877551078796387
    },
    {
      "auxiliary_loss_clip": 0.01164636,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.04897761,
      "balance_loss_mlp": 1.01980877,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.208851240210062,
      "language_loss": 0.64143527,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66335559,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.567894697189331
    },
    {
      "auxiliary_loss_clip": 0.01149335,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04330468,
      "balance_loss_mlp": 1.02028108,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 2.6635952823807894,
      "language_loss": 0.62240028,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.6441654,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 3.6295602321624756
    },
    {
      "auxiliary_loss_clip": 0.01150285,
      "auxiliary_loss_mlp": 0.01023738,
      "balance_loss_clip": 1.04445493,
      "balance_loss_mlp": 1.01677346,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 2.4477460016098247,
      "language_loss": 0.72100002,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74274027,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.6456263065338135
    },
    {
      "auxiliary_loss_clip": 0.01165072,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.04637539,
      "balance_loss_mlp": 1.01892376,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 1.812120680595449,
      "language_loss": 0.68839854,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71031296,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.689805507659912
    },
    {
      "auxiliary_loss_clip": 0.01158693,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04578149,
      "balance_loss_mlp": 1.01964688,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 2.2450809608313738,
      "language_loss": 0.76709551,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78895915,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 3.5689592361450195
    },
    {
      "auxiliary_loss_clip": 0.01153507,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04689407,
      "balance_loss_mlp": 1.01791596,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.1954641959657772,
      "language_loss": 0.8637712,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88555604,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.633054494857788
    },
    {
      "auxiliary_loss_clip": 0.01127009,
      "auxiliary_loss_mlp": 0.01019771,
      "balance_loss_clip": 1.04331207,
      "balance_loss_mlp": 1.01274967,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 2.007666430981588,
      "language_loss": 0.68057716,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70204496,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.7933738231658936
    },
    {
      "auxiliary_loss_clip": 0.01158637,
      "auxiliary_loss_mlp": 0.0102293,
      "balance_loss_clip": 1.04355586,
      "balance_loss_mlp": 1.01610529,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.40249385303316,
      "language_loss": 0.76163507,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78345072,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.6273903846740723
    },
    {
      "auxiliary_loss_clip": 0.01142664,
      "auxiliary_loss_mlp": 0.01034288,
      "balance_loss_clip": 1.04703903,
      "balance_loss_mlp": 1.02694798,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 2.1830199653146094,
      "language_loss": 0.7913959,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81316549,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.780177354812622
    },
    {
      "auxiliary_loss_clip": 0.01163999,
      "auxiliary_loss_mlp": 0.01022682,
      "balance_loss_clip": 1.04613698,
      "balance_loss_mlp": 1.01547241,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.694036149291354,
      "language_loss": 0.75484741,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77671421,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.683100700378418
    },
    {
      "auxiliary_loss_clip": 0.01156239,
      "auxiliary_loss_mlp": 0.01027049,
      "balance_loss_clip": 1.04933238,
      "balance_loss_mlp": 1.01983082,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.018288700397581,
      "language_loss": 0.75192481,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.7737577,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.6886708736419678
    },
    {
      "auxiliary_loss_clip": 0.01132654,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.04272127,
      "balance_loss_mlp": 1.01884127,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 1.8276311684003448,
      "language_loss": 0.72966099,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75124586,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 3.654102087020874
    },
    {
      "auxiliary_loss_clip": 0.01157579,
      "auxiliary_loss_mlp": 0.00901006,
      "balance_loss_clip": 1.048527,
      "balance_loss_mlp": 1.00081646,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.6191729646096378,
      "language_loss": 0.77779949,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.79838532,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.6173079013824463
    },
    {
      "auxiliary_loss_clip": 0.01148175,
      "auxiliary_loss_mlp": 0.0102451,
      "balance_loss_clip": 1.04431379,
      "balance_loss_mlp": 1.01694047,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.4146753117561186,
      "language_loss": 0.75305384,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77478063,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.6410634517669678
    },
    {
      "auxiliary_loss_clip": 0.01148774,
      "auxiliary_loss_mlp": 0.01022863,
      "balance_loss_clip": 1.04319525,
      "balance_loss_mlp": 1.01560891,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 2.194463333482334,
      "language_loss": 0.57047021,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59218657,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.741175413131714
    },
    {
      "auxiliary_loss_clip": 0.01063301,
      "auxiliary_loss_mlp": 0.01001779,
      "balance_loss_clip": 1.01209235,
      "balance_loss_mlp": 1.00100982,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6412618848350606,
      "language_loss": 0.51643527,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53708613,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.273413896560669
    },
    {
      "auxiliary_loss_clip": 0.01158968,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.0461297,
      "balance_loss_mlp": 1.0177443,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.8809210626558144,
      "language_loss": 0.88524687,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90708494,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.608778476715088
    },
    {
      "auxiliary_loss_clip": 0.01150503,
      "auxiliary_loss_mlp": 0.00900721,
      "balance_loss_clip": 1.0435462,
      "balance_loss_mlp": 1.00075388,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.4753583985440275,
      "language_loss": 0.78214538,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.8026576,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.6625101566314697
    },
    {
      "auxiliary_loss_clip": 0.01144017,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.04666114,
      "balance_loss_mlp": 1.02062249,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 3.9723363064184802,
      "language_loss": 0.66799498,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.68972003,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.766860008239746
    },
    {
      "auxiliary_loss_clip": 0.01141602,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.04679132,
      "balance_loss_mlp": 1.02010703,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.991273207867143,
      "language_loss": 0.68366933,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70536554,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.661038875579834
    },
    {
      "auxiliary_loss_clip": 0.0114294,
      "auxiliary_loss_mlp": 0.01021826,
      "balance_loss_clip": 1.04441667,
      "balance_loss_mlp": 1.01441753,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.2338747003420645,
      "language_loss": 0.80438268,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82603037,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 3.506443977355957
    },
    {
      "auxiliary_loss_clip": 0.0113264,
      "auxiliary_loss_mlp": 0.01028675,
      "balance_loss_clip": 1.04629493,
      "balance_loss_mlp": 1.0217638,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.0477682238616604,
      "language_loss": 0.68894041,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.71055353,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.7678349018096924
    },
    {
      "auxiliary_loss_clip": 0.01126128,
      "auxiliary_loss_mlp": 0.01029624,
      "balance_loss_clip": 1.04175258,
      "balance_loss_mlp": 1.02265072,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 2.291349604573404,
      "language_loss": 0.77965707,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.80121458,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.7418808937072754
    },
    {
      "auxiliary_loss_clip": 0.01153727,
      "auxiliary_loss_mlp": 0.01021806,
      "balance_loss_clip": 1.04660797,
      "balance_loss_mlp": 1.01465392,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.8060494236716749,
      "language_loss": 0.76163065,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78338605,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.7138261795043945
    },
    {
      "auxiliary_loss_clip": 0.01176585,
      "auxiliary_loss_mlp": 0.01025322,
      "balance_loss_clip": 1.05261791,
      "balance_loss_mlp": 1.01790094,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 2.251917951943138,
      "language_loss": 0.63795233,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.65997148,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.735325574874878
    },
    {
      "auxiliary_loss_clip": 0.01162663,
      "auxiliary_loss_mlp": 0.00900743,
      "balance_loss_clip": 1.04669356,
      "balance_loss_mlp": 1.00084949,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 3.0593595003037857,
      "language_loss": 0.5582608,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.57889485,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 2.681550979614258
    },
    {
      "auxiliary_loss_clip": 0.01165052,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04782581,
      "balance_loss_mlp": 1.01758254,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 2.2322942939142743,
      "language_loss": 0.74194187,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76384354,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.6695456504821777
    },
    {
      "auxiliary_loss_clip": 0.01153877,
      "auxiliary_loss_mlp": 0.01029595,
      "balance_loss_clip": 1.04597569,
      "balance_loss_mlp": 1.02235913,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.791702555898902,
      "language_loss": 0.75709754,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.77893227,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.6339058876037598
    },
    {
      "auxiliary_loss_clip": 0.01147461,
      "auxiliary_loss_mlp": 0.01023419,
      "balance_loss_clip": 1.04625034,
      "balance_loss_mlp": 1.01630282,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.7694207663621464,
      "language_loss": 0.65782332,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67953211,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 3.533616542816162
    },
    {
      "auxiliary_loss_clip": 0.01130887,
      "auxiliary_loss_mlp": 0.0090105,
      "balance_loss_clip": 1.04181981,
      "balance_loss_mlp": 1.00071037,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.3448554190672697,
      "language_loss": 0.76848924,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.78880858,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.745643377304077
    },
    {
      "auxiliary_loss_clip": 0.01150954,
      "auxiliary_loss_mlp": 0.01022592,
      "balance_loss_clip": 1.04564953,
      "balance_loss_mlp": 1.01553524,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.8167338811993494,
      "language_loss": 0.76111692,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.78285241,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 3.622366428375244
    },
    {
      "auxiliary_loss_clip": 0.01144417,
      "auxiliary_loss_mlp": 0.01021753,
      "balance_loss_clip": 1.04417634,
      "balance_loss_mlp": 1.01428413,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 2.3460536527960802,
      "language_loss": 0.67849588,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70015752,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.709489345550537
    },
    {
      "auxiliary_loss_clip": 0.0117179,
      "auxiliary_loss_mlp": 0.01026859,
      "balance_loss_clip": 1.0484848,
      "balance_loss_mlp": 1.01949787,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.245113914924176,
      "language_loss": 0.68307781,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70506436,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.5762767791748047
    },
    {
      "auxiliary_loss_clip": 0.01162493,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.04871655,
      "balance_loss_mlp": 1.01737857,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.9335053932489634,
      "language_loss": 0.84048426,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.86236155,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.666269540786743
    },
    {
      "auxiliary_loss_clip": 0.01160911,
      "auxiliary_loss_mlp": 0.01020526,
      "balance_loss_clip": 1.04601598,
      "balance_loss_mlp": 1.01274133,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 2.427144140248773,
      "language_loss": 0.82112563,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84294003,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.623061180114746
    },
    {
      "auxiliary_loss_clip": 0.01160651,
      "auxiliary_loss_mlp": 0.00900776,
      "balance_loss_clip": 1.04715228,
      "balance_loss_mlp": 1.00069332,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 2.8591806019909276,
      "language_loss": 0.73825711,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.75887138,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.565338611602783
    },
    {
      "auxiliary_loss_clip": 0.01146418,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04498005,
      "balance_loss_mlp": 1.01888275,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 2.4528189084042356,
      "language_loss": 0.77769297,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79942381,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.635603189468384
    },
    {
      "auxiliary_loss_clip": 0.01165665,
      "auxiliary_loss_mlp": 0.01026615,
      "balance_loss_clip": 1.0509491,
      "balance_loss_mlp": 1.01852608,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.8488488289541583,
      "language_loss": 0.76239139,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78431422,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.5566391944885254
    },
    {
      "auxiliary_loss_clip": 0.01161379,
      "auxiliary_loss_mlp": 0.01024277,
      "balance_loss_clip": 1.04800224,
      "balance_loss_mlp": 1.01714838,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 3.193873729018991,
      "language_loss": 0.71816593,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.74002242,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 3.5669102668762207
    },
    {
      "auxiliary_loss_clip": 0.01157832,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.04609036,
      "balance_loss_mlp": 1.01796436,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.139001269491949,
      "language_loss": 0.66888291,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.69071531,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.7032699584960938
    },
    {
      "auxiliary_loss_clip": 0.01159239,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.04573298,
      "balance_loss_mlp": 1.01712632,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 3.8695291469422606,
      "language_loss": 0.79922831,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82106483,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.6780519485473633
    },
    {
      "auxiliary_loss_clip": 0.01070737,
      "auxiliary_loss_mlp": 0.01000705,
      "balance_loss_clip": 1.0109973,
      "balance_loss_mlp": 0.99990034,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7746691795313202,
      "language_loss": 0.60054898,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62126344,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.2145490646362305
    },
    {
      "auxiliary_loss_clip": 0.01160289,
      "auxiliary_loss_mlp": 0.0102361,
      "balance_loss_clip": 1.04957771,
      "balance_loss_mlp": 1.01664543,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 2.361178113441678,
      "language_loss": 0.7389667,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.76080573,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.7167882919311523
    },
    {
      "auxiliary_loss_clip": 0.01153878,
      "auxiliary_loss_mlp": 0.01030611,
      "balance_loss_clip": 1.0443877,
      "balance_loss_mlp": 1.02236164,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.094653824816304,
      "language_loss": 0.79936612,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82121104,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.679779291152954
    },
    {
      "auxiliary_loss_clip": 0.01167332,
      "auxiliary_loss_mlp": 0.0102049,
      "balance_loss_clip": 1.04723299,
      "balance_loss_mlp": 1.01377296,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.266407072692478,
      "language_loss": 0.73361039,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75548863,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.5263166427612305
    },
    {
      "auxiliary_loss_clip": 0.01164787,
      "auxiliary_loss_mlp": 0.01027026,
      "balance_loss_clip": 1.04856396,
      "balance_loss_mlp": 1.01916718,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 2.6191097645952994,
      "language_loss": 0.79757285,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81949097,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.6620030403137207
    },
    {
      "auxiliary_loss_clip": 0.01136043,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.040627,
      "balance_loss_mlp": 1.01764059,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.8214106547436772,
      "language_loss": 0.66545403,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68707663,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.823298215866089
    },
    {
      "auxiliary_loss_clip": 0.01161764,
      "auxiliary_loss_mlp": 0.0102337,
      "balance_loss_clip": 1.04347563,
      "balance_loss_mlp": 1.01591325,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 2.036143367015788,
      "language_loss": 0.70278454,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72463584,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 3.58878493309021
    },
    {
      "auxiliary_loss_clip": 0.01164253,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.05019975,
      "balance_loss_mlp": 1.01543784,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.8858678475610073,
      "language_loss": 0.80960858,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.8314774,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.5366053581237793
    },
    {
      "auxiliary_loss_clip": 0.01145076,
      "auxiliary_loss_mlp": 0.01027331,
      "balance_loss_clip": 1.04431319,
      "balance_loss_mlp": 1.02050078,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 4.036638396552287,
      "language_loss": 0.73153114,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.75325525,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.5818428993225098
    },
    {
      "auxiliary_loss_clip": 0.01171912,
      "auxiliary_loss_mlp": 0.01027888,
      "balance_loss_clip": 1.04842615,
      "balance_loss_mlp": 1.01995504,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.7752054506613486,
      "language_loss": 0.70206153,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72405952,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.746255874633789
    },
    {
      "auxiliary_loss_clip": 0.01144404,
      "auxiliary_loss_mlp": 0.01029372,
      "balance_loss_clip": 1.04661083,
      "balance_loss_mlp": 1.02222228,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.8100435889012632,
      "language_loss": 0.73724496,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75898272,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.727055788040161
    },
    {
      "auxiliary_loss_clip": 0.01132351,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.04515982,
      "balance_loss_mlp": 1.01982689,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.097381298202525,
      "language_loss": 0.77644098,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.79803634,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.719041347503662
    },
    {
      "auxiliary_loss_clip": 0.01158991,
      "auxiliary_loss_mlp": 0.01022456,
      "balance_loss_clip": 1.04426932,
      "balance_loss_mlp": 1.01561356,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.5811878121729572,
      "language_loss": 0.71271336,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73452783,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 2.5603878498077393
    },
    {
      "auxiliary_loss_clip": 0.01150182,
      "auxiliary_loss_mlp": 0.01020489,
      "balance_loss_clip": 1.04612327,
      "balance_loss_mlp": 1.01260948,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 5.266273426697717,
      "language_loss": 0.67009735,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69180405,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.6273889541625977
    },
    {
      "auxiliary_loss_clip": 0.01131449,
      "auxiliary_loss_mlp": 0.01025911,
      "balance_loss_clip": 1.0442152,
      "balance_loss_mlp": 1.01819801,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 1.9743373859065745,
      "language_loss": 0.71131074,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73288435,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 3.587224245071411
    },
    {
      "auxiliary_loss_clip": 0.01150973,
      "auxiliary_loss_mlp": 0.01026293,
      "balance_loss_clip": 1.04550147,
      "balance_loss_mlp": 1.01920342,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 2.199839696554479,
      "language_loss": 0.68540579,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70717847,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 3.6050217151641846
    },
    {
      "auxiliary_loss_clip": 0.01151249,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04501534,
      "balance_loss_mlp": 1.01702297,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 3.019190999310659,
      "language_loss": 0.70465022,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72640824,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.6629416942596436
    },
    {
      "auxiliary_loss_clip": 0.01142416,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.04873157,
      "balance_loss_mlp": 1.01789141,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.8457982776678166,
      "language_loss": 0.87193179,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89360869,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.7366933822631836
    },
    {
      "auxiliary_loss_clip": 0.01160191,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 1.04463923,
      "balance_loss_mlp": 1.01663899,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 4.976545687405747,
      "language_loss": 0.77280813,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79464585,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.5783751010894775
    },
    {
      "auxiliary_loss_clip": 0.0116758,
      "auxiliary_loss_mlp": 0.010244,
      "balance_loss_clip": 1.04757643,
      "balance_loss_mlp": 1.01713383,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.767729179755334,
      "language_loss": 0.76393902,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78585887,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.581057548522949
    },
    {
      "auxiliary_loss_clip": 0.0115329,
      "auxiliary_loss_mlp": 0.01024696,
      "balance_loss_clip": 1.04516864,
      "balance_loss_mlp": 1.01732874,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.3801690684264205,
      "language_loss": 0.74346679,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76524663,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.7690820693969727
    },
    {
      "auxiliary_loss_clip": 0.01155237,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.04334998,
      "balance_loss_mlp": 1.01973772,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.809397235264083,
      "language_loss": 0.71502292,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73685253,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.6103410720825195
    },
    {
      "auxiliary_loss_clip": 0.01162241,
      "auxiliary_loss_mlp": 0.01021438,
      "balance_loss_clip": 1.04806113,
      "balance_loss_mlp": 1.01422882,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 2.380974756603783,
      "language_loss": 0.78367114,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.8055079,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.715601921081543
    },
    {
      "auxiliary_loss_clip": 0.01135882,
      "auxiliary_loss_mlp": 0.01024906,
      "balance_loss_clip": 1.04357767,
      "balance_loss_mlp": 1.01742244,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 4.303844211598154,
      "language_loss": 0.69515604,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71676391,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 3.5964252948760986
    },
    {
      "auxiliary_loss_clip": 0.01126905,
      "auxiliary_loss_mlp": 0.01022113,
      "balance_loss_clip": 1.04164433,
      "balance_loss_mlp": 1.01526785,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 3.367212931749197,
      "language_loss": 0.71335101,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.73484123,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.770113468170166
    },
    {
      "auxiliary_loss_clip": 0.01150383,
      "auxiliary_loss_mlp": 0.01026711,
      "balance_loss_clip": 1.04769742,
      "balance_loss_mlp": 1.01899242,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 2.1064136256593002,
      "language_loss": 0.77653408,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.79830498,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.680603504180908
    },
    {
      "auxiliary_loss_clip": 0.01120041,
      "auxiliary_loss_mlp": 0.01025493,
      "balance_loss_clip": 1.03466845,
      "balance_loss_mlp": 1.01747954,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.186840111493436,
      "language_loss": 0.68116653,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70262182,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.7539405822753906
    },
    {
      "auxiliary_loss_clip": 0.01153736,
      "auxiliary_loss_mlp": 0.01025016,
      "balance_loss_clip": 1.04998207,
      "balance_loss_mlp": 1.01707363,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 1.8817643861283375,
      "language_loss": 0.71527231,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73705983,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.6869165897369385
    },
    {
      "auxiliary_loss_clip": 0.01135021,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.04212236,
      "balance_loss_mlp": 1.01902092,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.7272896901458052,
      "language_loss": 0.71928525,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.74090052,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.6485493183135986
    },
    {
      "auxiliary_loss_clip": 0.01154996,
      "auxiliary_loss_mlp": 0.01027698,
      "balance_loss_clip": 1.04499853,
      "balance_loss_mlp": 1.02039337,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.4414585141406953,
      "language_loss": 0.72769248,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74951941,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.8177402019500732
    },
    {
      "auxiliary_loss_clip": 0.01153599,
      "auxiliary_loss_mlp": 0.01027842,
      "balance_loss_clip": 1.04695106,
      "balance_loss_mlp": 1.02006936,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.666063753752599,
      "language_loss": 0.69772458,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.71953899,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.6627304553985596
    },
    {
      "auxiliary_loss_clip": 0.01163979,
      "auxiliary_loss_mlp": 0.01024166,
      "balance_loss_clip": 1.05004954,
      "balance_loss_mlp": 1.01645601,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.8806934428921087,
      "language_loss": 0.84358639,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86546779,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.6136820316314697
    },
    {
      "auxiliary_loss_clip": 0.0114294,
      "auxiliary_loss_mlp": 0.01022642,
      "balance_loss_clip": 1.04466915,
      "balance_loss_mlp": 1.01532257,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 2.3545908981628725,
      "language_loss": 0.73048234,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75213814,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.7879221439361572
    },
    {
      "auxiliary_loss_clip": 0.01144548,
      "auxiliary_loss_mlp": 0.01022043,
      "balance_loss_clip": 1.04539084,
      "balance_loss_mlp": 1.01538777,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 2.2351841331771785,
      "language_loss": 0.6506089,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67227477,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 3.6937124729156494
    },
    {
      "auxiliary_loss_clip": 0.01160556,
      "auxiliary_loss_mlp": 0.01022492,
      "balance_loss_clip": 1.04500651,
      "balance_loss_mlp": 1.01495528,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.852804661445696,
      "language_loss": 0.72410876,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74593925,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.6371965408325195
    },
    {
      "auxiliary_loss_clip": 0.01166799,
      "auxiliary_loss_mlp": 0.01028285,
      "balance_loss_clip": 1.04719436,
      "balance_loss_mlp": 1.02045846,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.7653833192374355,
      "language_loss": 0.63758957,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.65954041,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.541166067123413
    },
    {
      "auxiliary_loss_clip": 0.01161286,
      "auxiliary_loss_mlp": 0.01024311,
      "balance_loss_clip": 1.04772508,
      "balance_loss_mlp": 1.01630652,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.844409250996513,
      "language_loss": 0.60991561,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.63177156,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 2.7002193927764893
    },
    {
      "auxiliary_loss_clip": 0.01132065,
      "auxiliary_loss_mlp": 0.01025524,
      "balance_loss_clip": 1.04628229,
      "balance_loss_mlp": 1.01794279,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.4209118879354676,
      "language_loss": 0.76040709,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78198302,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.654920816421509
    },
    {
      "auxiliary_loss_clip": 0.01150556,
      "auxiliary_loss_mlp": 0.00900709,
      "balance_loss_clip": 1.04360747,
      "balance_loss_mlp": 1.00074768,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.6835622587175598,
      "language_loss": 0.87144399,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89195669,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.5957934856414795
    },
    {
      "auxiliary_loss_clip": 0.01153105,
      "auxiliary_loss_mlp": 0.01026161,
      "balance_loss_clip": 1.04754639,
      "balance_loss_mlp": 1.01846004,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 2.164912927133636,
      "language_loss": 0.64921725,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.6710099,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.604344367980957
    },
    {
      "auxiliary_loss_clip": 0.01134281,
      "auxiliary_loss_mlp": 0.01022878,
      "balance_loss_clip": 1.04717934,
      "balance_loss_mlp": 1.01577878,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.7616897253332693,
      "language_loss": 0.71302152,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73459309,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 3.674649477005005
    },
    {
      "auxiliary_loss_clip": 0.01170034,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.04748809,
      "balance_loss_mlp": 1.02077937,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 2.3882944621732802,
      "language_loss": 0.79620731,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.81818992,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.6106903553009033
    },
    {
      "auxiliary_loss_clip": 0.01154031,
      "auxiliary_loss_mlp": 0.01027522,
      "balance_loss_clip": 1.04419017,
      "balance_loss_mlp": 1.0202738,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.5960501057455794,
      "language_loss": 0.77159584,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79341137,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 3.5551655292510986
    },
    {
      "auxiliary_loss_clip": 0.01151666,
      "auxiliary_loss_mlp": 0.01024364,
      "balance_loss_clip": 1.04496527,
      "balance_loss_mlp": 1.0175842,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.733872810929528,
      "language_loss": 0.82876456,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.8505249,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.5803422927856445
    },
    {
      "auxiliary_loss_clip": 0.0112283,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04225302,
      "balance_loss_mlp": 1.02022934,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 2.195949654292572,
      "language_loss": 0.74091679,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76241922,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.726547956466675
    },
    {
      "auxiliary_loss_clip": 0.01142403,
      "auxiliary_loss_mlp": 0.01022212,
      "balance_loss_clip": 1.04145789,
      "balance_loss_mlp": 1.01509571,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 2.0551707600546614,
      "language_loss": 0.73870367,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76034975,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.655183792114258
    },
    {
      "auxiliary_loss_clip": 0.01168606,
      "auxiliary_loss_mlp": 0.01021991,
      "balance_loss_clip": 1.04870105,
      "balance_loss_mlp": 1.01476109,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.9517767893733073,
      "language_loss": 0.80671453,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.82862055,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.538125991821289
    },
    {
      "auxiliary_loss_clip": 0.01160629,
      "auxiliary_loss_mlp": 0.01024918,
      "balance_loss_clip": 1.04596698,
      "balance_loss_mlp": 1.01685929,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.806191511827411,
      "language_loss": 0.77277553,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79463089,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.6038002967834473
    },
    {
      "auxiliary_loss_clip": 0.01168535,
      "auxiliary_loss_mlp": 0.01024094,
      "balance_loss_clip": 1.0493741,
      "balance_loss_mlp": 1.01727581,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 2.0315427966476483,
      "language_loss": 0.7352069,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75713319,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.5546112060546875
    },
    {
      "auxiliary_loss_clip": 0.01137622,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.04617608,
      "balance_loss_mlp": 1.02055931,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.7004052626759365,
      "language_loss": 0.82336688,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84501898,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.7675554752349854
    },
    {
      "auxiliary_loss_clip": 0.01129007,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.04077911,
      "balance_loss_mlp": 1.01824284,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 2.410561478996126,
      "language_loss": 0.71416569,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.7357052,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 3.6815879344940186
    },
    {
      "auxiliary_loss_clip": 0.01154061,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.04537845,
      "balance_loss_mlp": 1.01936674,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 3.6355367297437255,
      "language_loss": 0.61416698,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63598341,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.6823832988739014
    },
    {
      "auxiliary_loss_clip": 0.01171462,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.04776967,
      "balance_loss_mlp": 1.01957178,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.082029765621763,
      "language_loss": 0.843521,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.8655076,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.5347185134887695
    },
    {
      "auxiliary_loss_clip": 0.01137331,
      "auxiliary_loss_mlp": 0.01021271,
      "balance_loss_clip": 1.04254007,
      "balance_loss_mlp": 1.01458955,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.5516995908671323,
      "language_loss": 0.83248997,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85407603,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.784496545791626
    },
    {
      "auxiliary_loss_clip": 0.01154024,
      "auxiliary_loss_mlp": 0.01021877,
      "balance_loss_clip": 1.04666591,
      "balance_loss_mlp": 1.01429534,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.640863712332688,
      "language_loss": 0.86838222,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.89014119,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.6399247646331787
    },
    {
      "auxiliary_loss_clip": 0.01163419,
      "auxiliary_loss_mlp": 0.00900567,
      "balance_loss_clip": 1.04846191,
      "balance_loss_mlp": 1.00073767,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.3399777739023486,
      "language_loss": 0.73086464,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.75150448,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.6511733531951904
    },
    {
      "auxiliary_loss_clip": 0.01143995,
      "auxiliary_loss_mlp": 0.01022549,
      "balance_loss_clip": 1.04416752,
      "balance_loss_mlp": 1.01489592,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.4157184198955854,
      "language_loss": 0.86932302,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89098847,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.6492345333099365
    },
    {
      "auxiliary_loss_clip": 0.01168708,
      "auxiliary_loss_mlp": 0.01023264,
      "balance_loss_clip": 1.04908156,
      "balance_loss_mlp": 1.01622498,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 1.92510376371799,
      "language_loss": 0.63110226,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.65302199,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.6113481521606445
    },
    {
      "auxiliary_loss_clip": 0.01149176,
      "auxiliary_loss_mlp": 0.01024796,
      "balance_loss_clip": 1.04440939,
      "balance_loss_mlp": 1.01799774,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 2.3217357844818807,
      "language_loss": 0.79277217,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.81451184,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.6537277698516846
    },
    {
      "auxiliary_loss_clip": 0.0115153,
      "auxiliary_loss_mlp": 0.01025422,
      "balance_loss_clip": 1.05123115,
      "balance_loss_mlp": 1.01793599,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.4584161543234773,
      "language_loss": 0.75781363,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.7795831,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 3.5251622200012207
    },
    {
      "auxiliary_loss_clip": 0.01175756,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.05156136,
      "balance_loss_mlp": 1.018996,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 2.046594108439887,
      "language_loss": 0.75693798,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77896249,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.5488975048065186
    },
    {
      "auxiliary_loss_clip": 0.01143354,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 1.04292202,
      "balance_loss_mlp": 1.02019799,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.841797679893942,
      "language_loss": 0.70068055,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.72238779,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.675016164779663
    },
    {
      "auxiliary_loss_clip": 0.01169801,
      "auxiliary_loss_mlp": 0.01025629,
      "balance_loss_clip": 1.0468744,
      "balance_loss_mlp": 1.01818442,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.733642444257803,
      "language_loss": 0.77521443,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79716873,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 2.5295825004577637
    },
    {
      "auxiliary_loss_clip": 0.01172808,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 1.04988694,
      "balance_loss_mlp": 1.01866722,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.6502908552010127,
      "language_loss": 0.78562158,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80760896,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.538081645965576
    },
    {
      "auxiliary_loss_clip": 0.01148337,
      "auxiliary_loss_mlp": 0.01025195,
      "balance_loss_clip": 1.04226887,
      "balance_loss_mlp": 1.01752377,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.977158779582115,
      "language_loss": 0.69329441,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71502972,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.6056854724884033
    },
    {
      "auxiliary_loss_clip": 0.01126527,
      "auxiliary_loss_mlp": 0.0102197,
      "balance_loss_clip": 1.03770208,
      "balance_loss_mlp": 1.01441169,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.698420903340415,
      "language_loss": 0.74097514,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76246017,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.713181257247925
    },
    {
      "auxiliary_loss_clip": 0.01152892,
      "auxiliary_loss_mlp": 0.01022404,
      "balance_loss_clip": 1.04418254,
      "balance_loss_mlp": 1.01464915,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 2.5875653317222898,
      "language_loss": 0.77447397,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79622698,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 3.6348798274993896
    },
    {
      "auxiliary_loss_clip": 0.01145995,
      "auxiliary_loss_mlp": 0.01032274,
      "balance_loss_clip": 1.04494095,
      "balance_loss_mlp": 1.02463245,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.2286085413606913,
      "language_loss": 0.78829944,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81008208,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.618826150894165
    },
    {
      "auxiliary_loss_clip": 0.01157513,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.04505348,
      "balance_loss_mlp": 1.02334511,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 2.552535066482251,
      "language_loss": 0.63072723,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65261477,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 3.5854945182800293
    },
    {
      "auxiliary_loss_clip": 0.01151674,
      "auxiliary_loss_mlp": 0.01024804,
      "balance_loss_clip": 1.04608214,
      "balance_loss_mlp": 1.01758313,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 1.8629608310149561,
      "language_loss": 0.85520065,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87696546,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.61964750289917
    },
    {
      "auxiliary_loss_clip": 0.01155847,
      "auxiliary_loss_mlp": 0.01023484,
      "balance_loss_clip": 1.04492342,
      "balance_loss_mlp": 1.01590192,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.002694869813816,
      "language_loss": 0.75875986,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78055322,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.6826529502868652
    },
    {
      "auxiliary_loss_clip": 0.01156117,
      "auxiliary_loss_mlp": 0.01023191,
      "balance_loss_clip": 1.04664469,
      "balance_loss_mlp": 1.01646495,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.8926120619446767,
      "language_loss": 0.80303347,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82482654,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.6324872970581055
    },
    {
      "auxiliary_loss_clip": 0.01151355,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04532886,
      "balance_loss_mlp": 1.01743412,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.2386309716415425,
      "language_loss": 0.83735216,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.85910678,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.607922315597534
    },
    {
      "auxiliary_loss_clip": 0.01165023,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.0482651,
      "balance_loss_mlp": 1.01921225,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.772847574922431,
      "language_loss": 0.81206858,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83398795,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.58974027633667
    },
    {
      "auxiliary_loss_clip": 0.01145112,
      "auxiliary_loss_mlp": 0.01027921,
      "balance_loss_clip": 1.04879498,
      "balance_loss_mlp": 1.02008891,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 1.7891086337244824,
      "language_loss": 0.81122243,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83295274,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.755772113800049
    },
    {
      "auxiliary_loss_clip": 0.01147624,
      "auxiliary_loss_mlp": 0.01027697,
      "balance_loss_clip": 1.04306173,
      "balance_loss_mlp": 1.02018976,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.7369437719023066,
      "language_loss": 0.68785053,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.70960373,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.619638442993164
    },
    {
      "auxiliary_loss_clip": 0.01166142,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.05053937,
      "balance_loss_mlp": 1.01926279,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 2.211957241490908,
      "language_loss": 0.86921346,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.8911463,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.6665992736816406
    },
    {
      "auxiliary_loss_clip": 0.01133613,
      "auxiliary_loss_mlp": 0.01027474,
      "balance_loss_clip": 1.03840089,
      "balance_loss_mlp": 1.02010727,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.7480639604054162,
      "language_loss": 0.7700696,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.79168051,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 3.6963870525360107
    },
    {
      "auxiliary_loss_clip": 0.01154179,
      "auxiliary_loss_mlp": 0.01023818,
      "balance_loss_clip": 1.04556274,
      "balance_loss_mlp": 1.01674557,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 2.323851817094873,
      "language_loss": 0.78819221,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.80997223,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.6891796588897705
    },
    {
      "auxiliary_loss_clip": 0.01138922,
      "auxiliary_loss_mlp": 0.01026676,
      "balance_loss_clip": 1.04467809,
      "balance_loss_mlp": 1.01978886,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 2.112407905257871,
      "language_loss": 0.6638118,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68546784,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.645092010498047
    },
    {
      "auxiliary_loss_clip": 0.01169218,
      "auxiliary_loss_mlp": 0.00901689,
      "balance_loss_clip": 1.0483036,
      "balance_loss_mlp": 1.00078642,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 2.0020423123129474,
      "language_loss": 0.71991342,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.74062252,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.5451292991638184
    },
    {
      "auxiliary_loss_clip": 0.01142244,
      "auxiliary_loss_mlp": 0.010287,
      "balance_loss_clip": 1.04444075,
      "balance_loss_mlp": 1.021065,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 2.6212873465967763,
      "language_loss": 0.68915641,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71086586,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.707750082015991
    },
    {
      "auxiliary_loss_clip": 0.01135414,
      "auxiliary_loss_mlp": 0.00901378,
      "balance_loss_clip": 1.04130054,
      "balance_loss_mlp": 1.00072026,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.7018114705438958,
      "language_loss": 0.77896571,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.79933363,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.6675353050231934
    },
    {
      "auxiliary_loss_clip": 0.01153498,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.04658651,
      "balance_loss_mlp": 1.02347183,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.2371556508605033,
      "language_loss": 0.62272888,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64457834,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.713240385055542
    },
    {
      "auxiliary_loss_clip": 0.01167391,
      "auxiliary_loss_mlp": 0.01024789,
      "balance_loss_clip": 1.04622924,
      "balance_loss_mlp": 1.01753521,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.1317230876013773,
      "language_loss": 0.6159867,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63790846,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.6642391681671143
    },
    {
      "auxiliary_loss_clip": 0.01123633,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.04202366,
      "balance_loss_mlp": 1.02040005,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.5900510960272811,
      "language_loss": 0.7787891,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80029827,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.7907416820526123
    },
    {
      "auxiliary_loss_clip": 0.01130822,
      "auxiliary_loss_mlp": 0.01021652,
      "balance_loss_clip": 1.03948951,
      "balance_loss_mlp": 1.01454413,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.7413439122262662,
      "language_loss": 0.83830112,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.85982585,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 3.7109103202819824
    },
    {
      "auxiliary_loss_clip": 0.0115977,
      "auxiliary_loss_mlp": 0.01023853,
      "balance_loss_clip": 1.04842687,
      "balance_loss_mlp": 1.01625919,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 2.095685145473123,
      "language_loss": 0.77415335,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.79598951,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.5540835857391357
    },
    {
      "auxiliary_loss_clip": 0.01178373,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.05339766,
      "balance_loss_mlp": 1.01866388,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 2.8780832746184943,
      "language_loss": 0.91376269,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93581265,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.5622780323028564
    },
    {
      "auxiliary_loss_clip": 0.01125077,
      "auxiliary_loss_mlp": 0.01025232,
      "balance_loss_clip": 1.04568064,
      "balance_loss_mlp": 1.01777875,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.4710888321938067,
      "language_loss": 0.86985844,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89136147,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 2.6814229488372803
    },
    {
      "auxiliary_loss_clip": 0.01163632,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.05016041,
      "balance_loss_mlp": 1.0161314,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 3.6597634247994115,
      "language_loss": 0.80752861,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.82940251,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.570362091064453
    },
    {
      "auxiliary_loss_clip": 0.01171339,
      "auxiliary_loss_mlp": 0.01024982,
      "balance_loss_clip": 1.04886794,
      "balance_loss_mlp": 1.01744175,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.9184844325560286,
      "language_loss": 0.78328568,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80524886,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.601369857788086
    },
    {
      "auxiliary_loss_clip": 0.01151297,
      "auxiliary_loss_mlp": 0.01021326,
      "balance_loss_clip": 1.04864323,
      "balance_loss_mlp": 1.014081,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 8.228198289982261,
      "language_loss": 0.74012446,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76185066,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.6823678016662598
    },
    {
      "auxiliary_loss_clip": 0.01163911,
      "auxiliary_loss_mlp": 0.01026355,
      "balance_loss_clip": 1.04636395,
      "balance_loss_mlp": 1.01925635,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 1.9531552016691203,
      "language_loss": 0.66087711,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68277979,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 3.6021177768707275
    },
    {
      "auxiliary_loss_clip": 0.0116422,
      "auxiliary_loss_mlp": 0.00901245,
      "balance_loss_clip": 1.04676926,
      "balance_loss_mlp": 1.00084591,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 1.7428453764076193,
      "language_loss": 0.7983619,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81901652,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.633384943008423
    },
    {
      "auxiliary_loss_clip": 0.01066738,
      "auxiliary_loss_mlp": 0.01000741,
      "balance_loss_clip": 1.01086283,
      "balance_loss_mlp": 1.00003147,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.9178635348796157,
      "language_loss": 0.59518778,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61586255,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 4.055598258972168
    },
    {
      "auxiliary_loss_clip": 0.01119405,
      "auxiliary_loss_mlp": 0.01024228,
      "balance_loss_clip": 1.03991103,
      "balance_loss_mlp": 1.01714659,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 9.80610611029015,
      "language_loss": 0.78852928,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.80996561,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.8150415420532227
    },
    {
      "auxiliary_loss_clip": 0.01169982,
      "auxiliary_loss_mlp": 0.01026335,
      "balance_loss_clip": 1.04870939,
      "balance_loss_mlp": 1.01908433,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.758636927030726,
      "language_loss": 0.66315353,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.68511665,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.5303475856781006
    },
    {
      "auxiliary_loss_clip": 0.01122656,
      "auxiliary_loss_mlp": 0.00901844,
      "balance_loss_clip": 1.04201603,
      "balance_loss_mlp": 1.00069654,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.2923944620469943,
      "language_loss": 0.7804426,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80068755,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.816538095474243
    },
    {
      "auxiliary_loss_clip": 0.01149339,
      "auxiliary_loss_mlp": 0.01029307,
      "balance_loss_clip": 1.04602504,
      "balance_loss_mlp": 1.02190661,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 2.1151323374497317,
      "language_loss": 0.74816847,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.76995504,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.6356537342071533
    },
    {
      "auxiliary_loss_clip": 0.01155351,
      "auxiliary_loss_mlp": 0.01028281,
      "balance_loss_clip": 1.04506516,
      "balance_loss_mlp": 1.02093124,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.9259998940443095,
      "language_loss": 0.74351752,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.7653538,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.621572732925415
    },
    {
      "auxiliary_loss_clip": 0.01164553,
      "auxiliary_loss_mlp": 0.01021599,
      "balance_loss_clip": 1.05082154,
      "balance_loss_mlp": 1.01407707,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 3.6467324761828297,
      "language_loss": 0.81127572,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83313721,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.559197187423706
    },
    {
      "auxiliary_loss_clip": 0.01137848,
      "auxiliary_loss_mlp": 0.01029156,
      "balance_loss_clip": 1.04549646,
      "balance_loss_mlp": 1.02161646,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 2.2190389644387403,
      "language_loss": 0.81593478,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83760488,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.7014377117156982
    },
    {
      "auxiliary_loss_clip": 0.01159481,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 1.04554367,
      "balance_loss_mlp": 1.0183059,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 2.052747627981964,
      "language_loss": 0.74112701,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76297998,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 3.526723623275757
    },
    {
      "auxiliary_loss_clip": 0.01126463,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.03734422,
      "balance_loss_mlp": 1.0195601,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.772113578095396,
      "language_loss": 0.7657733,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78730613,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.7836833000183105
    },
    {
      "auxiliary_loss_clip": 0.01158802,
      "auxiliary_loss_mlp": 0.01025413,
      "balance_loss_clip": 1.04425812,
      "balance_loss_mlp": 1.01851654,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.8522466417838406,
      "language_loss": 0.78489637,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80673856,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.6441197395324707
    },
    {
      "auxiliary_loss_clip": 0.011539,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.04688466,
      "balance_loss_mlp": 1.0213362,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 3.40059860257486,
      "language_loss": 0.78548372,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80731148,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.6476497650146484
    },
    {
      "auxiliary_loss_clip": 0.01169497,
      "auxiliary_loss_mlp": 0.01027054,
      "balance_loss_clip": 1.04774022,
      "balance_loss_mlp": 1.01919818,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.5009789815798253,
      "language_loss": 0.73940897,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76137447,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.6356067657470703
    },
    {
      "auxiliary_loss_clip": 0.01153184,
      "auxiliary_loss_mlp": 0.01025737,
      "balance_loss_clip": 1.0436511,
      "balance_loss_mlp": 1.01848876,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.4228532687213082,
      "language_loss": 0.74692065,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76870984,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.6590702533721924
    },
    {
      "auxiliary_loss_clip": 0.01161299,
      "auxiliary_loss_mlp": 0.01024857,
      "balance_loss_clip": 1.05126977,
      "balance_loss_mlp": 1.01757908,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.773502206317144,
      "language_loss": 0.7922132,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81407475,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.691054105758667
    },
    {
      "auxiliary_loss_clip": 0.01171319,
      "auxiliary_loss_mlp": 0.01021763,
      "balance_loss_clip": 1.0493753,
      "balance_loss_mlp": 1.013901,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.0117517369684057,
      "language_loss": 0.81078929,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83272004,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.578749418258667
    },
    {
      "auxiliary_loss_clip": 0.01139965,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.04269063,
      "balance_loss_mlp": 1.01993883,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 1.9527808553162889,
      "language_loss": 0.8293066,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.85097969,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.6981406211853027
    },
    {
      "auxiliary_loss_clip": 0.01066723,
      "auxiliary_loss_mlp": 0.01001026,
      "balance_loss_clip": 1.01065779,
      "balance_loss_mlp": 1.00015604,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7338510404629887,
      "language_loss": 0.61866093,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63933837,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 4.175726652145386
    },
    {
      "auxiliary_loss_clip": 0.01148226,
      "auxiliary_loss_mlp": 0.01028267,
      "balance_loss_clip": 1.04865217,
      "balance_loss_mlp": 1.020262,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 2.125391877118474,
      "language_loss": 0.76311034,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78487527,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.670552968978882
    },
    {
      "auxiliary_loss_clip": 0.01135213,
      "auxiliary_loss_mlp": 0.01024211,
      "balance_loss_clip": 1.04245138,
      "balance_loss_mlp": 1.01732635,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.7353824905370727,
      "language_loss": 0.82439065,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84598494,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.7659595012664795
    },
    {
      "auxiliary_loss_clip": 0.0115478,
      "auxiliary_loss_mlp": 0.01025518,
      "balance_loss_clip": 1.04487634,
      "balance_loss_mlp": 1.01852679,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 1.7433767813321093,
      "language_loss": 0.62190318,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.6437062,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.598766326904297
    },
    {
      "auxiliary_loss_clip": 0.0113306,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.03642082,
      "balance_loss_mlp": 1.01844347,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.285747953700708,
      "language_loss": 0.7062071,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72779793,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.7474002838134766
    },
    {
      "auxiliary_loss_clip": 0.01165321,
      "auxiliary_loss_mlp": 0.01023637,
      "balance_loss_clip": 1.04826045,
      "balance_loss_mlp": 1.01661539,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 2.1525815964465744,
      "language_loss": 0.79219967,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81408924,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.6416101455688477
    },
    {
      "auxiliary_loss_clip": 0.01150301,
      "auxiliary_loss_mlp": 0.00901462,
      "balance_loss_clip": 1.04493582,
      "balance_loss_mlp": 1.00076818,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 3.5106339770514836,
      "language_loss": 0.6959303,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71644795,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.7628672122955322
    },
    {
      "auxiliary_loss_clip": 0.0113748,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.0442636,
      "balance_loss_mlp": 1.01911926,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.5650496856498837,
      "language_loss": 0.6100899,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63173592,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 3.617868423461914
    },
    {
      "auxiliary_loss_clip": 0.01139266,
      "auxiliary_loss_mlp": 0.01031857,
      "balance_loss_clip": 1.04282081,
      "balance_loss_mlp": 1.02389956,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.7593750531748944,
      "language_loss": 0.76568449,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.78739572,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.7152106761932373
    },
    {
      "auxiliary_loss_clip": 0.01147053,
      "auxiliary_loss_mlp": 0.01023829,
      "balance_loss_clip": 1.04392433,
      "balance_loss_mlp": 1.0155623,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.2376385982560785,
      "language_loss": 0.89199436,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.9137032,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 3.591625452041626
    },
    {
      "auxiliary_loss_clip": 0.01154481,
      "auxiliary_loss_mlp": 0.0102392,
      "balance_loss_clip": 1.04735589,
      "balance_loss_mlp": 1.01690722,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 3.3433030126823606,
      "language_loss": 0.76915979,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79094386,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.643796443939209
    },
    {
      "auxiliary_loss_clip": 0.01161818,
      "auxiliary_loss_mlp": 0.01023533,
      "balance_loss_clip": 1.04713833,
      "balance_loss_mlp": 1.01660681,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 1.996911951511596,
      "language_loss": 0.79665422,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.81850767,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.605618953704834
    },
    {
      "auxiliary_loss_clip": 0.01168409,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.04867637,
      "balance_loss_mlp": 1.01919401,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 1.9692126891710635,
      "language_loss": 0.81528866,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83723438,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.531493902206421
    },
    {
      "auxiliary_loss_clip": 0.01149008,
      "auxiliary_loss_mlp": 0.01033655,
      "balance_loss_clip": 1.0440979,
      "balance_loss_mlp": 1.02542973,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 3.2671691226901953,
      "language_loss": 0.71254534,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73437202,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.7838573455810547
    },
    {
      "auxiliary_loss_clip": 0.01162793,
      "auxiliary_loss_mlp": 0.01022981,
      "balance_loss_clip": 1.0486176,
      "balance_loss_mlp": 1.01554298,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.8367982074251987,
      "language_loss": 0.70672226,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.72858006,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.566572666168213
    },
    {
      "auxiliary_loss_clip": 0.01155682,
      "auxiliary_loss_mlp": 0.01022821,
      "balance_loss_clip": 1.04409504,
      "balance_loss_mlp": 1.01534617,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.7305534331473889,
      "language_loss": 0.85174155,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87352657,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.6181132793426514
    },
    {
      "auxiliary_loss_clip": 0.01167923,
      "auxiliary_loss_mlp": 0.01024713,
      "balance_loss_clip": 1.04573894,
      "balance_loss_mlp": 1.01744151,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 2.071934898154532,
      "language_loss": 0.79177618,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81370252,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.5794291496276855
    },
    {
      "auxiliary_loss_clip": 0.01167976,
      "auxiliary_loss_mlp": 0.0102364,
      "balance_loss_clip": 1.04776859,
      "balance_loss_mlp": 1.01627874,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.9342181553976554,
      "language_loss": 0.77455616,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79647231,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 3.4892430305480957
    },
    {
      "auxiliary_loss_clip": 0.01162224,
      "auxiliary_loss_mlp": 0.01026969,
      "balance_loss_clip": 1.04521024,
      "balance_loss_mlp": 1.01964653,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.8867507218470914,
      "language_loss": 0.75658631,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77847826,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.647965908050537
    },
    {
      "auxiliary_loss_clip": 0.0115069,
      "auxiliary_loss_mlp": 0.01023885,
      "balance_loss_clip": 1.04586864,
      "balance_loss_mlp": 1.01681018,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 1.9503018446050666,
      "language_loss": 0.80847871,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.83022445,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.58217453956604
    },
    {
      "auxiliary_loss_clip": 0.0116976,
      "auxiliary_loss_mlp": 0.01022806,
      "balance_loss_clip": 1.04830456,
      "balance_loss_mlp": 1.01500416,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 3.343493586870467,
      "language_loss": 0.81805032,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.83997601,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.5686492919921875
    },
    {
      "auxiliary_loss_clip": 0.01131529,
      "auxiliary_loss_mlp": 0.01022247,
      "balance_loss_clip": 1.04081237,
      "balance_loss_mlp": 1.01510668,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 2.2367042076665324,
      "language_loss": 0.83206511,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.85360295,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.7218923568725586
    },
    {
      "auxiliary_loss_clip": 0.01152049,
      "auxiliary_loss_mlp": 0.01023319,
      "balance_loss_clip": 1.04366851,
      "balance_loss_mlp": 1.01646996,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 2.14303533205575,
      "language_loss": 0.70967185,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73142552,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.6494946479797363
    },
    {
      "auxiliary_loss_clip": 0.01129667,
      "auxiliary_loss_mlp": 0.01027134,
      "balance_loss_clip": 1.04192352,
      "balance_loss_mlp": 1.01992178,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.20489954885274,
      "language_loss": 0.83258164,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85414958,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.7038190364837646
    },
    {
      "auxiliary_loss_clip": 0.01170998,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.0493108,
      "balance_loss_mlp": 1.01861405,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 2.464166593960947,
      "language_loss": 0.76831943,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.79029298,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.540860414505005
    },
    {
      "auxiliary_loss_clip": 0.01148495,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.04438996,
      "balance_loss_mlp": 1.01998234,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.0338963156565764,
      "language_loss": 0.67973566,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.70149601,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.566248893737793
    },
    {
      "auxiliary_loss_clip": 0.01149944,
      "auxiliary_loss_mlp": 0.01027835,
      "balance_loss_clip": 1.04618037,
      "balance_loss_mlp": 1.02028012,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 1.8917038734799705,
      "language_loss": 0.7790994,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80087721,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 3.51812744140625
    },
    {
      "auxiliary_loss_clip": 0.01150709,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04279542,
      "balance_loss_mlp": 1.01695299,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.8147031633290918,
      "language_loss": 0.74449074,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76624537,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.5649640560150146
    },
    {
      "auxiliary_loss_clip": 0.01065386,
      "auxiliary_loss_mlp": 0.0100075,
      "balance_loss_clip": 1.01186347,
      "balance_loss_mlp": 0.99994487,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7519347576664315,
      "language_loss": 0.5695672,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59022856,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.127581834793091
    },
    {
      "auxiliary_loss_clip": 0.01143423,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.04239464,
      "balance_loss_mlp": 1.02050698,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 1.9448196024385411,
      "language_loss": 0.68542695,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70714194,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 2.7800958156585693
    },
    {
      "auxiliary_loss_clip": 0.01145819,
      "auxiliary_loss_mlp": 0.01019843,
      "balance_loss_clip": 1.04470861,
      "balance_loss_mlp": 1.01288462,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.638622194269403,
      "language_loss": 0.80227745,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82393408,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.635326862335205
    },
    {
      "auxiliary_loss_clip": 0.0116105,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.04855871,
      "balance_loss_mlp": 1.01770806,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 2.003149644300383,
      "language_loss": 0.72834295,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.75020504,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.587984800338745
    },
    {
      "auxiliary_loss_clip": 0.01158617,
      "auxiliary_loss_mlp": 0.01028329,
      "balance_loss_clip": 1.04957283,
      "balance_loss_mlp": 1.0206579,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 2.2870723817270564,
      "language_loss": 0.76351964,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78538907,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.5974555015563965
    },
    {
      "auxiliary_loss_clip": 0.01116672,
      "auxiliary_loss_mlp": 0.01022625,
      "balance_loss_clip": 1.03981423,
      "balance_loss_mlp": 1.01440525,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.48809620229451,
      "language_loss": 0.7671386,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78853154,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 3.697896957397461
    },
    {
      "auxiliary_loss_clip": 0.01050937,
      "auxiliary_loss_mlp": 0.01001888,
      "balance_loss_clip": 1.01259089,
      "balance_loss_mlp": 1.00100613,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.721482663305626,
      "language_loss": 0.5507403,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57126856,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.1400389671325684
    },
    {
      "auxiliary_loss_clip": 0.01152548,
      "auxiliary_loss_mlp": 0.01028086,
      "balance_loss_clip": 1.04542232,
      "balance_loss_mlp": 1.02007198,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.15849616738799,
      "language_loss": 0.69993365,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72174001,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.6412417888641357
    },
    {
      "auxiliary_loss_clip": 0.01059396,
      "auxiliary_loss_mlp": 0.01002835,
      "balance_loss_clip": 1.01344156,
      "balance_loss_mlp": 1.00202489,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8390767992610891,
      "language_loss": 0.62003338,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64065564,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 4.064562082290649
    },
    {
      "auxiliary_loss_clip": 0.0116169,
      "auxiliary_loss_mlp": 0.01023867,
      "balance_loss_clip": 1.04699278,
      "balance_loss_mlp": 1.01643479,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.605374599411326,
      "language_loss": 0.76076663,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.78262222,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.694641351699829
    },
    {
      "auxiliary_loss_clip": 0.01171122,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.04815316,
      "balance_loss_mlp": 1.02006996,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.2350879735354714,
      "language_loss": 0.82296836,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84495467,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.526181936264038
    },
    {
      "auxiliary_loss_clip": 0.01156182,
      "auxiliary_loss_mlp": 0.01021948,
      "balance_loss_clip": 1.04577208,
      "balance_loss_mlp": 1.01515341,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 2.02049831022768,
      "language_loss": 0.70628929,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.72807068,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.59438157081604
    },
    {
      "auxiliary_loss_clip": 0.011117,
      "auxiliary_loss_mlp": 0.01024268,
      "balance_loss_clip": 1.03910196,
      "balance_loss_mlp": 1.01668954,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.960790991919538,
      "language_loss": 0.80618966,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.8275494,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.75789475440979
    },
    {
      "auxiliary_loss_clip": 0.01152554,
      "auxiliary_loss_mlp": 0.01020818,
      "balance_loss_clip": 1.047822,
      "balance_loss_mlp": 1.01338553,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 5.7310000257294265,
      "language_loss": 0.76050591,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78223968,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.6000733375549316
    },
    {
      "auxiliary_loss_clip": 0.01154417,
      "auxiliary_loss_mlp": 0.0103474,
      "balance_loss_clip": 1.04857826,
      "balance_loss_mlp": 1.02712297,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 2.0658915450601256,
      "language_loss": 0.70891035,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.73080188,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.6271932125091553
    },
    {
      "auxiliary_loss_clip": 0.01161814,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04841208,
      "balance_loss_mlp": 1.01750112,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 1.991681820018776,
      "language_loss": 0.77541,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79727846,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 3.5248606204986572
    },
    {
      "auxiliary_loss_clip": 0.01162038,
      "auxiliary_loss_mlp": 0.0102516,
      "balance_loss_clip": 1.0510819,
      "balance_loss_mlp": 1.01759648,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.6885019508382682,
      "language_loss": 0.78590018,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.80777216,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.6272685527801514
    },
    {
      "auxiliary_loss_clip": 0.01127335,
      "auxiliary_loss_mlp": 0.01027879,
      "balance_loss_clip": 1.04115188,
      "balance_loss_mlp": 1.02008891,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.3945674205193495,
      "language_loss": 0.82092232,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84247446,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.763502836227417
    },
    {
      "auxiliary_loss_clip": 0.01130748,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.04072142,
      "balance_loss_mlp": 1.01834786,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.6745993502311889,
      "language_loss": 0.75453401,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.7760973,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.719363212585449
    },
    {
      "auxiliary_loss_clip": 0.01165881,
      "auxiliary_loss_mlp": 0.0102296,
      "balance_loss_clip": 1.04667735,
      "balance_loss_mlp": 1.0161736,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.0337797676596865,
      "language_loss": 0.78145927,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80334771,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.5538063049316406
    },
    {
      "auxiliary_loss_clip": 0.01162328,
      "auxiliary_loss_mlp": 0.01032649,
      "balance_loss_clip": 1.0480212,
      "balance_loss_mlp": 1.02423882,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.1072819480498515,
      "language_loss": 0.81225908,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83420885,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.6148433685302734
    },
    {
      "auxiliary_loss_clip": 0.01164964,
      "auxiliary_loss_mlp": 0.01022657,
      "balance_loss_clip": 1.04808629,
      "balance_loss_mlp": 1.01553726,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 2.631310211017372,
      "language_loss": 0.69277948,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.7146557,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.6939291954040527
    },
    {
      "auxiliary_loss_clip": 0.01160106,
      "auxiliary_loss_mlp": 0.01025698,
      "balance_loss_clip": 1.04765797,
      "balance_loss_mlp": 1.01876569,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.224339021556769,
      "language_loss": 0.73006511,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.75192314,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.6202163696289062
    },
    {
      "auxiliary_loss_clip": 0.01159514,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.04573488,
      "balance_loss_mlp": 1.01733959,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 2.4111831378269724,
      "language_loss": 0.73525482,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.7570926,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.634092092514038
    },
    {
      "auxiliary_loss_clip": 0.01170889,
      "auxiliary_loss_mlp": 0.00900873,
      "balance_loss_clip": 1.04912829,
      "balance_loss_mlp": 1.0008471,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 3.70325266278551,
      "language_loss": 0.78210866,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80282629,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.532376766204834
    },
    {
      "auxiliary_loss_clip": 0.01172152,
      "auxiliary_loss_mlp": 0.01021827,
      "balance_loss_clip": 1.04905367,
      "balance_loss_mlp": 1.01448381,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 2.0752689628634253,
      "language_loss": 0.77462173,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.7965616,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 3.4855117797851562
    },
    {
      "auxiliary_loss_clip": 0.01059941,
      "auxiliary_loss_mlp": 0.01002099,
      "balance_loss_clip": 1.01201069,
      "balance_loss_mlp": 1.00129414,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.7217784067342828,
      "language_loss": 0.65493166,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67555201,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 3.2354350090026855
    },
    {
      "auxiliary_loss_clip": 0.01169457,
      "auxiliary_loss_mlp": 0.01026316,
      "balance_loss_clip": 1.04683971,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 2.1586240943835318,
      "language_loss": 0.73640579,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.75836349,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.5783016681671143
    },
    {
      "auxiliary_loss_clip": 0.0116766,
      "auxiliary_loss_mlp": 0.0102327,
      "balance_loss_clip": 1.0459094,
      "balance_loss_mlp": 1.01592696,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 2.9988012173642202,
      "language_loss": 0.67568237,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.69759166,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 2.772089719772339
    },
    {
      "auxiliary_loss_clip": 0.01164949,
      "auxiliary_loss_mlp": 0.00901153,
      "balance_loss_clip": 1.04643893,
      "balance_loss_mlp": 1.00075161,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 2.799224074677874,
      "language_loss": 0.7759636,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79662466,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.667435646057129
    },
    {
      "auxiliary_loss_clip": 0.0117255,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.04946351,
      "balance_loss_mlp": 1.02038097,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 4.683149184868789,
      "language_loss": 0.70059216,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72260541,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.5340054035186768
    },
    {
      "auxiliary_loss_clip": 0.01171524,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.0468837,
      "balance_loss_mlp": 1.01882768,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 2.1566883160373362,
      "language_loss": 0.77594531,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79792428,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 2.6420724391937256
    },
    {
      "auxiliary_loss_clip": 0.01173988,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.05009317,
      "balance_loss_mlp": 1.01771927,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.7314736537628805,
      "language_loss": 0.67088723,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.69288087,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 3.5171284675598145
    },
    {
      "auxiliary_loss_clip": 0.01154192,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.04483616,
      "balance_loss_mlp": 1.0161947,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.9800884335899611,
      "language_loss": 0.69654906,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71832895,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.5937349796295166
    },
    {
      "auxiliary_loss_clip": 0.01154876,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.04587996,
      "balance_loss_mlp": 1.01640534,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 2.2961555149943984,
      "language_loss": 0.78738147,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.80916971,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.6024301052093506
    },
    {
      "auxiliary_loss_clip": 0.01066885,
      "auxiliary_loss_mlp": 0.01002268,
      "balance_loss_clip": 1.01102781,
      "balance_loss_mlp": 1.00145185,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8012520844281055,
      "language_loss": 0.64714026,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66783178,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 4.022964239120483
    },
    {
      "auxiliary_loss_clip": 0.01143759,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.04609084,
      "balance_loss_mlp": 1.01620638,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 2.035386035227468,
      "language_loss": 0.74790287,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76956689,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.616995334625244
    },
    {
      "auxiliary_loss_clip": 0.01148133,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.0430336,
      "balance_loss_mlp": 1.01950121,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.2349945088841436,
      "language_loss": 0.65246302,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67421246,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.6165695190429688
    },
    {
      "auxiliary_loss_clip": 0.01142413,
      "auxiliary_loss_mlp": 0.01024879,
      "balance_loss_clip": 1.04528379,
      "balance_loss_mlp": 1.01751184,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.8211790546939521,
      "language_loss": 0.78299612,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80466908,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.7094948291778564
    },
    {
      "auxiliary_loss_clip": 0.01158778,
      "auxiliary_loss_mlp": 0.00900807,
      "balance_loss_clip": 1.04609382,
      "balance_loss_mlp": 1.00071287,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.565206034484774,
      "language_loss": 0.74365008,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.76424599,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.5993735790252686
    },
    {
      "auxiliary_loss_clip": 0.01168663,
      "auxiliary_loss_mlp": 0.01027041,
      "balance_loss_clip": 1.04914641,
      "balance_loss_mlp": 1.01931059,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 1.9098001889185585,
      "language_loss": 0.66298008,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68493712,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.6353342533111572
    },
    {
      "auxiliary_loss_clip": 0.01173816,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.04840517,
      "balance_loss_mlp": 1.02071834,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 2.494297340802413,
      "language_loss": 0.68093014,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.7029562,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.6841893196105957
    },
    {
      "auxiliary_loss_clip": 0.01173731,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.0519284,
      "balance_loss_mlp": 1.02005935,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.7892189999094905,
      "language_loss": 0.7333858,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75539464,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 3.618385076522827
    },
    {
      "auxiliary_loss_clip": 0.01142979,
      "auxiliary_loss_mlp": 0.0102671,
      "balance_loss_clip": 1.04443169,
      "balance_loss_mlp": 1.01948881,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.284935554583371,
      "language_loss": 0.74356872,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76526564,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.6967456340789795
    },
    {
      "auxiliary_loss_clip": 0.01169286,
      "auxiliary_loss_mlp": 0.0102493,
      "balance_loss_clip": 1.04786336,
      "balance_loss_mlp": 1.01769078,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 2.0812453028150544,
      "language_loss": 0.74534678,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.76728892,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.5485007762908936
    },
    {
      "auxiliary_loss_clip": 0.01138878,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.0427289,
      "balance_loss_mlp": 1.01842344,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 3.367952598215113,
      "language_loss": 0.79401362,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81565696,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.7333409786224365
    },
    {
      "auxiliary_loss_clip": 0.0117211,
      "auxiliary_loss_mlp": 0.01021814,
      "balance_loss_clip": 1.04893839,
      "balance_loss_mlp": 1.01438177,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 2.121113454034225,
      "language_loss": 0.78919524,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.81113452,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.617631196975708
    },
    {
      "auxiliary_loss_clip": 0.01160208,
      "auxiliary_loss_mlp": 0.01034899,
      "balance_loss_clip": 1.04880953,
      "balance_loss_mlp": 1.0269444,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 4.410809594166974,
      "language_loss": 0.71825981,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.74021089,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.6857004165649414
    },
    {
      "auxiliary_loss_clip": 0.01172175,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.04959488,
      "balance_loss_mlp": 1.01987624,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 3.4466655674559052,
      "language_loss": 0.77546489,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79746127,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.529904842376709
    },
    {
      "auxiliary_loss_clip": 0.01159381,
      "auxiliary_loss_mlp": 0.01022204,
      "balance_loss_clip": 1.04819751,
      "balance_loss_mlp": 1.01562035,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.9508739901457952,
      "language_loss": 0.74460322,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76641905,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.766512393951416
    },
    {
      "auxiliary_loss_clip": 0.0114942,
      "auxiliary_loss_mlp": 0.01024415,
      "balance_loss_clip": 1.04572439,
      "balance_loss_mlp": 1.01642215,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 6.214005754094561,
      "language_loss": 0.6514492,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67318761,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.6782398223876953
    },
    {
      "auxiliary_loss_clip": 0.01170815,
      "auxiliary_loss_mlp": 0.01024448,
      "balance_loss_clip": 1.04854357,
      "balance_loss_mlp": 1.0167979,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.9051233328355264,
      "language_loss": 0.73814631,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.76009887,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.607870101928711
    },
    {
      "auxiliary_loss_clip": 0.0112791,
      "auxiliary_loss_mlp": 0.0102451,
      "balance_loss_clip": 1.04179084,
      "balance_loss_mlp": 1.01706815,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.797512599606311,
      "language_loss": 0.73986495,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76138914,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 3.7019906044006348
    },
    {
      "auxiliary_loss_clip": 0.01166609,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.05054152,
      "balance_loss_mlp": 1.01995063,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.119049965927158,
      "language_loss": 0.74224949,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76419508,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.6048779487609863
    },
    {
      "auxiliary_loss_clip": 0.01136542,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.04030633,
      "balance_loss_mlp": 1.01964355,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 2.1274429895182374,
      "language_loss": 0.74085248,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76248646,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 2.6405649185180664
    },
    {
      "auxiliary_loss_clip": 0.01162394,
      "auxiliary_loss_mlp": 0.01022025,
      "balance_loss_clip": 1.04631114,
      "balance_loss_mlp": 1.01484311,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.662559853883638,
      "language_loss": 0.79398572,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81582987,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.5605709552764893
    },
    {
      "auxiliary_loss_clip": 0.01151511,
      "auxiliary_loss_mlp": 0.01027042,
      "balance_loss_clip": 1.04441464,
      "balance_loss_mlp": 1.01957929,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.201532740269773,
      "language_loss": 0.76139784,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.7831834,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.6342456340789795
    },
    {
      "auxiliary_loss_clip": 0.01047434,
      "auxiliary_loss_mlp": 0.01001762,
      "balance_loss_clip": 1.00912428,
      "balance_loss_mlp": 1.00091541,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.8585261619444258,
      "language_loss": 0.6079911,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62848306,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.1893458366394043
    },
    {
      "auxiliary_loss_clip": 0.01168235,
      "auxiliary_loss_mlp": 0.01025682,
      "balance_loss_clip": 1.04815888,
      "balance_loss_mlp": 1.01865435,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 2.1128617345123875,
      "language_loss": 0.80412781,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82606697,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.5902810096740723
    },
    {
      "auxiliary_loss_clip": 0.01152097,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04564667,
      "balance_loss_mlp": 1.01785541,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.0205097088904234,
      "language_loss": 0.77009279,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.79186124,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 3.6164779663085938
    },
    {
      "auxiliary_loss_clip": 0.0113294,
      "auxiliary_loss_mlp": 0.00901492,
      "balance_loss_clip": 1.04446507,
      "balance_loss_mlp": 1.00073183,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.6834061459140175,
      "language_loss": 0.75968933,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78003365,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.766909122467041
    },
    {
      "auxiliary_loss_clip": 0.01151756,
      "auxiliary_loss_mlp": 0.01024003,
      "balance_loss_clip": 1.04661489,
      "balance_loss_mlp": 1.01690423,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.8146005328275698,
      "language_loss": 0.83973098,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86148858,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 3.5196828842163086
    },
    {
      "auxiliary_loss_clip": 0.01146352,
      "auxiliary_loss_mlp": 0.01021933,
      "balance_loss_clip": 1.04810512,
      "balance_loss_mlp": 1.01458955,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.6248291642265404,
      "language_loss": 0.72810602,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.74978882,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.628295421600342
    },
    {
      "auxiliary_loss_clip": 0.01153591,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.04590189,
      "balance_loss_mlp": 1.01530266,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.8040515681142124,
      "language_loss": 0.70110393,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72287375,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.6894543170928955
    },
    {
      "auxiliary_loss_clip": 0.0115767,
      "auxiliary_loss_mlp": 0.01022102,
      "balance_loss_clip": 1.04684377,
      "balance_loss_mlp": 1.01500034,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 3.2202594448481934,
      "language_loss": 0.68662226,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70841998,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.65193772315979
    },
    {
      "auxiliary_loss_clip": 0.01135149,
      "auxiliary_loss_mlp": 0.01025098,
      "balance_loss_clip": 1.03882205,
      "balance_loss_mlp": 1.01715255,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 2.140451641235737,
      "language_loss": 0.84632075,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.8679232,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.705167293548584
    },
    {
      "auxiliary_loss_clip": 0.01164275,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.04827309,
      "balance_loss_mlp": 1.01687789,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 2.817483686653374,
      "language_loss": 0.82618606,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84807134,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.586806535720825
    },
    {
      "auxiliary_loss_clip": 0.01160818,
      "auxiliary_loss_mlp": 0.01022739,
      "balance_loss_clip": 1.04911208,
      "balance_loss_mlp": 1.01523232,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.7511571484811674,
      "language_loss": 0.77518475,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79702038,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.605088472366333
    },
    {
      "auxiliary_loss_clip": 0.01146338,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.04423857,
      "balance_loss_mlp": 1.01801324,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 1.9728541403115287,
      "language_loss": 0.7418139,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76353884,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.6710569858551025
    },
    {
      "auxiliary_loss_clip": 0.01168997,
      "auxiliary_loss_mlp": 0.01023643,
      "balance_loss_clip": 1.04775059,
      "balance_loss_mlp": 1.01649022,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 2.1257854720336127,
      "language_loss": 0.76426077,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78618717,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 3.4451615810394287
    },
    {
      "auxiliary_loss_clip": 0.01148061,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04487646,
      "balance_loss_mlp": 1.01455307,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 12.586654636537231,
      "language_loss": 0.64430356,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.6660074,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.693899154663086
    },
    {
      "auxiliary_loss_clip": 0.01170543,
      "auxiliary_loss_mlp": 0.0102458,
      "balance_loss_clip": 1.04999495,
      "balance_loss_mlp": 1.01742721,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 3.534792976732294,
      "language_loss": 0.852534,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87448525,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.569213390350342
    },
    {
      "auxiliary_loss_clip": 0.01160477,
      "auxiliary_loss_mlp": 0.00901151,
      "balance_loss_clip": 1.04746103,
      "balance_loss_mlp": 1.00080824,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 1.9365519295863027,
      "language_loss": 0.81402987,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83464622,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.6288206577301025
    },
    {
      "auxiliary_loss_clip": 0.01155927,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.05235255,
      "balance_loss_mlp": 1.01868176,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.839064518190571,
      "language_loss": 0.75742626,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77925205,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.6021456718444824
    },
    {
      "auxiliary_loss_clip": 0.01137215,
      "auxiliary_loss_mlp": 0.01016323,
      "balance_loss_clip": 1.04470468,
      "balance_loss_mlp": 1.00960207,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.866117958920388,
      "language_loss": 0.70527136,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72680676,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.7097184658050537
    },
    {
      "auxiliary_loss_clip": 0.01148728,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.0449369,
      "balance_loss_mlp": 1.02085376,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5858466989731839,
      "language_loss": 0.72200894,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74377608,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.710284471511841
    },
    {
      "auxiliary_loss_clip": 0.01165606,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.04906952,
      "balance_loss_mlp": 1.02169871,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 3.8904960236336605,
      "language_loss": 0.77578676,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79773819,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.8373336791992188
    },
    {
      "auxiliary_loss_clip": 0.01155619,
      "auxiliary_loss_mlp": 0.01025305,
      "balance_loss_clip": 1.04682291,
      "balance_loss_mlp": 1.01867056,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.068073978745117,
      "language_loss": 0.78885108,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.8106603,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.6668012142181396
    },
    {
      "auxiliary_loss_clip": 0.01151192,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 1.04285073,
      "balance_loss_mlp": 1.01795697,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.9470108705011326,
      "language_loss": 0.795766,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81754136,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 3.5480759143829346
    },
    {
      "auxiliary_loss_clip": 0.01141536,
      "auxiliary_loss_mlp": 0.01024055,
      "balance_loss_clip": 1.04564619,
      "balance_loss_mlp": 1.01642609,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.5424448307893637,
      "language_loss": 0.80309743,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.8247534,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.7780377864837646
    },
    {
      "auxiliary_loss_clip": 0.0115578,
      "auxiliary_loss_mlp": 0.01020857,
      "balance_loss_clip": 1.04598975,
      "balance_loss_mlp": 1.01378191,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.5659182805464043,
      "language_loss": 0.80825102,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.83001739,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.6031599044799805
    },
    {
      "auxiliary_loss_clip": 0.01173726,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05097294,
      "balance_loss_mlp": 1.01943254,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 3.6724361780550123,
      "language_loss": 0.75682521,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.77883089,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.5845229625701904
    },
    {
      "auxiliary_loss_clip": 0.01070268,
      "auxiliary_loss_mlp": 0.00890952,
      "balance_loss_clip": 1.01063633,
      "balance_loss_mlp": 0.99995941,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8281681718883535,
      "language_loss": 0.63925087,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65886307,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.2991631031036377
    },
    {
      "auxiliary_loss_clip": 0.01134048,
      "auxiliary_loss_mlp": 0.01019768,
      "balance_loss_clip": 1.04334772,
      "balance_loss_mlp": 1.01332796,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.783401014238189,
      "language_loss": 0.70029503,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72183323,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 2.690948247909546
    },
    {
      "auxiliary_loss_clip": 0.01149136,
      "auxiliary_loss_mlp": 0.01025264,
      "balance_loss_clip": 1.04538214,
      "balance_loss_mlp": 1.01815641,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.7922357760225012,
      "language_loss": 0.64452595,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.6662699,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.7815098762512207
    },
    {
      "auxiliary_loss_clip": 0.01169076,
      "auxiliary_loss_mlp": 0.00900847,
      "balance_loss_clip": 1.04821146,
      "balance_loss_mlp": 1.00079823,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 2.34858901150806,
      "language_loss": 0.74444151,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76514077,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 3.5382752418518066
    },
    {
      "auxiliary_loss_clip": 0.01157471,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04225636,
      "balance_loss_mlp": 1.01660573,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.7096508491985205,
      "language_loss": 0.72466981,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.7464934,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.7338647842407227
    },
    {
      "auxiliary_loss_clip": 0.01138919,
      "auxiliary_loss_mlp": 0.01025838,
      "balance_loss_clip": 1.04870224,
      "balance_loss_mlp": 1.01766598,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.6720364403351482,
      "language_loss": 0.80983841,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.83148599,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.654979944229126
    },
    {
      "auxiliary_loss_clip": 0.01147171,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.04305911,
      "balance_loss_mlp": 1.0193789,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.759302900356423,
      "language_loss": 0.63577366,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.65751362,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 3.648129940032959
    },
    {
      "auxiliary_loss_clip": 0.01142371,
      "auxiliary_loss_mlp": 0.01021397,
      "balance_loss_clip": 1.04558086,
      "balance_loss_mlp": 1.01443481,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 2.0409704332216974,
      "language_loss": 0.70588803,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72752571,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.764259099960327
    },
    {
      "auxiliary_loss_clip": 0.01046462,
      "auxiliary_loss_mlp": 0.01004461,
      "balance_loss_clip": 1.01073694,
      "balance_loss_mlp": 1.00345409,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7308649976309034,
      "language_loss": 0.57168359,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59219283,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.2459027767181396
    },
    {
      "auxiliary_loss_clip": 0.01051399,
      "auxiliary_loss_mlp": 0.01004566,
      "balance_loss_clip": 1.01074076,
      "balance_loss_mlp": 1.00369,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.8159510576297555,
      "language_loss": 0.58965826,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61021793,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.1137120723724365
    },
    {
      "auxiliary_loss_clip": 0.01153124,
      "auxiliary_loss_mlp": 0.01026752,
      "balance_loss_clip": 1.04549754,
      "balance_loss_mlp": 1.01899743,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.6518667139177676,
      "language_loss": 0.77782601,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79962474,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.601016044616699
    },
    {
      "auxiliary_loss_clip": 0.01154719,
      "auxiliary_loss_mlp": 0.01022712,
      "balance_loss_clip": 1.04778886,
      "balance_loss_mlp": 1.01535416,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 2.004291629307357,
      "language_loss": 0.76406884,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78584319,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.6703193187713623
    },
    {
      "auxiliary_loss_clip": 0.01156703,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.04370809,
      "balance_loss_mlp": 1.01714015,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.485325688351559,
      "language_loss": 0.65455377,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.6763674,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.614147663116455
    },
    {
      "auxiliary_loss_clip": 0.01172272,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.05060494,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 2.160022251787256,
      "language_loss": 0.72734296,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74933761,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 3.563836097717285
    },
    {
      "auxiliary_loss_clip": 0.01149352,
      "auxiliary_loss_mlp": 0.01021888,
      "balance_loss_clip": 1.04324651,
      "balance_loss_mlp": 1.01492286,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.7379149589925627,
      "language_loss": 0.74897528,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.7706877,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.713956117630005
    },
    {
      "auxiliary_loss_clip": 0.0117149,
      "auxiliary_loss_mlp": 0.01022113,
      "balance_loss_clip": 1.0489049,
      "balance_loss_mlp": 1.01507354,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.8741115275504634,
      "language_loss": 0.78726113,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.80919719,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.577270984649658
    },
    {
      "auxiliary_loss_clip": 0.01147783,
      "auxiliary_loss_mlp": 0.01025323,
      "balance_loss_clip": 1.04669213,
      "balance_loss_mlp": 1.01824808,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 1.8484750497042508,
      "language_loss": 0.75797886,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.77970994,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.7411937713623047
    },
    {
      "auxiliary_loss_clip": 0.01164886,
      "auxiliary_loss_mlp": 0.0102422,
      "balance_loss_clip": 1.04855776,
      "balance_loss_mlp": 1.01748443,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.3842461943900513,
      "language_loss": 0.83553123,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.85742229,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.5917720794677734
    },
    {
      "auxiliary_loss_clip": 0.01149852,
      "auxiliary_loss_mlp": 0.01024623,
      "balance_loss_clip": 1.04544306,
      "balance_loss_mlp": 1.01711631,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 1.7405461236821975,
      "language_loss": 0.79415739,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81590211,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.687368392944336
    },
    {
      "auxiliary_loss_clip": 0.01148757,
      "auxiliary_loss_mlp": 0.01019524,
      "balance_loss_clip": 1.04322815,
      "balance_loss_mlp": 1.01206136,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 3.3747431902443776,
      "language_loss": 0.64889354,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67057633,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.667489767074585
    },
    {
      "auxiliary_loss_clip": 0.01157359,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04479671,
      "balance_loss_mlp": 1.01717532,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 2.0147455108299237,
      "language_loss": 0.80126888,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.8230902,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.646603584289551
    },
    {
      "auxiliary_loss_clip": 0.01154393,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.04286599,
      "balance_loss_mlp": 1.01723218,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 2.3688404637534073,
      "language_loss": 0.7286129,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75040162,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.625746011734009
    },
    {
      "auxiliary_loss_clip": 0.01151454,
      "auxiliary_loss_mlp": 0.01028492,
      "balance_loss_clip": 1.04603457,
      "balance_loss_mlp": 1.02040339,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 1.9042655365934802,
      "language_loss": 0.87593818,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89773762,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 3.565887689590454
    },
    {
      "auxiliary_loss_clip": 0.01162074,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.04467189,
      "balance_loss_mlp": 1.01686835,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.565872618879357,
      "language_loss": 0.70753121,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.7293939,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.685603141784668
    },
    {
      "auxiliary_loss_clip": 0.01146409,
      "auxiliary_loss_mlp": 0.01020796,
      "balance_loss_clip": 1.045627,
      "balance_loss_mlp": 1.01381612,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 2.1661333057161176,
      "language_loss": 0.78677237,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.80844438,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.6405529975891113
    },
    {
      "auxiliary_loss_clip": 0.01138563,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.04532111,
      "balance_loss_mlp": 1.01987576,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 1.8264884955138023,
      "language_loss": 0.71545953,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73712224,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.7518510818481445
    },
    {
      "auxiliary_loss_clip": 0.01136626,
      "auxiliary_loss_mlp": 0.01023211,
      "balance_loss_clip": 1.04184794,
      "balance_loss_mlp": 1.01636815,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 2.0887281515766563,
      "language_loss": 0.64242494,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66402334,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.626945972442627
    },
    {
      "auxiliary_loss_clip": 0.01162548,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 1.05044651,
      "balance_loss_mlp": 1.01879823,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 3.4262737945038726,
      "language_loss": 0.82884979,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.85073721,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.613370656967163
    },
    {
      "auxiliary_loss_clip": 0.01137483,
      "auxiliary_loss_mlp": 0.01023253,
      "balance_loss_clip": 1.04494011,
      "balance_loss_mlp": 1.01590979,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.9192124999684663,
      "language_loss": 0.78964925,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81125665,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.7008938789367676
    },
    {
      "auxiliary_loss_clip": 0.01168478,
      "auxiliary_loss_mlp": 0.01025033,
      "balance_loss_clip": 1.04871798,
      "balance_loss_mlp": 1.01743352,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.9822348251403308,
      "language_loss": 0.71212184,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73405695,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.6981356143951416
    },
    {
      "auxiliary_loss_clip": 0.01159026,
      "auxiliary_loss_mlp": 0.01024375,
      "balance_loss_clip": 1.04611325,
      "balance_loss_mlp": 1.01706195,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 2.008403585002291,
      "language_loss": 0.85142726,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87326127,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 3.5513696670532227
    },
    {
      "auxiliary_loss_clip": 0.01160662,
      "auxiliary_loss_mlp": 0.0090156,
      "balance_loss_clip": 1.04772091,
      "balance_loss_mlp": 1.00090194,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 2.394636313671914,
      "language_loss": 0.68336427,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70398647,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.6537351608276367
    },
    {
      "auxiliary_loss_clip": 0.01154429,
      "auxiliary_loss_mlp": 0.01030291,
      "balance_loss_clip": 1.04769742,
      "balance_loss_mlp": 1.02238178,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.4033452631732715,
      "language_loss": 0.69675589,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.71860307,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.6724538803100586
    },
    {
      "auxiliary_loss_clip": 0.01151018,
      "auxiliary_loss_mlp": 0.01024732,
      "balance_loss_clip": 1.04960263,
      "balance_loss_mlp": 1.01721537,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.9385785269547606,
      "language_loss": 0.78796351,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.80972105,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 3.5281825065612793
    },
    {
      "auxiliary_loss_clip": 0.01159375,
      "auxiliary_loss_mlp": 0.009013,
      "balance_loss_clip": 1.04523349,
      "balance_loss_mlp": 1.00093889,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.0463980164122777,
      "language_loss": 0.78404176,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.80464852,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.6406779289245605
    },
    {
      "auxiliary_loss_clip": 0.01169354,
      "auxiliary_loss_mlp": 0.01023344,
      "balance_loss_clip": 1.04803717,
      "balance_loss_mlp": 1.01590514,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.491060755485075,
      "language_loss": 0.78316462,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80509162,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.5097239017486572
    },
    {
      "auxiliary_loss_clip": 0.01165022,
      "auxiliary_loss_mlp": 0.01023893,
      "balance_loss_clip": 1.04906607,
      "balance_loss_mlp": 1.01604319,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 5.095100942016509,
      "language_loss": 0.78656566,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80845481,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.544274091720581
    },
    {
      "auxiliary_loss_clip": 0.01165562,
      "auxiliary_loss_mlp": 0.01024713,
      "balance_loss_clip": 1.04510748,
      "balance_loss_mlp": 1.01736403,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.2103805247314563,
      "language_loss": 0.7286672,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75056994,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.564804792404175
    },
    {
      "auxiliary_loss_clip": 0.01172013,
      "auxiliary_loss_mlp": 0.01029532,
      "balance_loss_clip": 1.04880738,
      "balance_loss_mlp": 1.02196825,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.4451786739412487,
      "language_loss": 0.64827812,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67029357,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 2.9341816902160645
    },
    {
      "auxiliary_loss_clip": 0.01171249,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.05071664,
      "balance_loss_mlp": 1.01869011,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 1.9983096080863367,
      "language_loss": 0.78849268,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.8104651,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.5939905643463135
    },
    {
      "auxiliary_loss_clip": 0.01156221,
      "auxiliary_loss_mlp": 0.01020391,
      "balance_loss_clip": 1.04682446,
      "balance_loss_mlp": 1.01318538,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.1538949095894138,
      "language_loss": 0.72036123,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74212736,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 3.5822927951812744
    },
    {
      "auxiliary_loss_clip": 0.01146803,
      "auxiliary_loss_mlp": 0.01027734,
      "balance_loss_clip": 1.04737306,
      "balance_loss_mlp": 1.0202831,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.706054735725605,
      "language_loss": 0.74077374,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.76251912,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.721235513687134
    },
    {
      "auxiliary_loss_clip": 0.01154753,
      "auxiliary_loss_mlp": 0.01022867,
      "balance_loss_clip": 1.04458618,
      "balance_loss_mlp": 1.0156517,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.2935187927121117,
      "language_loss": 0.67641658,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69819283,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.612607479095459
    },
    {
      "auxiliary_loss_clip": 0.01168992,
      "auxiliary_loss_mlp": 0.01022151,
      "balance_loss_clip": 1.04929996,
      "balance_loss_mlp": 1.01505184,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 2.7213121289568636,
      "language_loss": 0.7486949,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.77060628,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.63535475730896
    },
    {
      "auxiliary_loss_clip": 0.0116517,
      "auxiliary_loss_mlp": 0.01020873,
      "balance_loss_clip": 1.04975033,
      "balance_loss_mlp": 1.01394415,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 1.79899310760493,
      "language_loss": 0.7398808,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76174122,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.610520839691162
    },
    {
      "auxiliary_loss_clip": 0.01165412,
      "auxiliary_loss_mlp": 0.01026087,
      "balance_loss_clip": 1.04798162,
      "balance_loss_mlp": 1.01817143,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 1.9605560277120402,
      "language_loss": 0.73135489,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75326985,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.6768651008605957
    },
    {
      "auxiliary_loss_clip": 0.01152512,
      "auxiliary_loss_mlp": 0.01021029,
      "balance_loss_clip": 1.04544282,
      "balance_loss_mlp": 1.01392388,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 2.2581522689811275,
      "language_loss": 0.6778568,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69959223,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.725184917449951
    },
    {
      "auxiliary_loss_clip": 0.01154279,
      "auxiliary_loss_mlp": 0.00901355,
      "balance_loss_clip": 1.04938805,
      "balance_loss_mlp": 1.00076473,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 4.746940767645681,
      "language_loss": 0.85265177,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87320817,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.7888760566711426
    },
    {
      "auxiliary_loss_clip": 0.01163791,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.04802752,
      "balance_loss_mlp": 1.01996052,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 2.096432310924636,
      "language_loss": 0.81929874,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84121931,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.740845203399658
    },
    {
      "auxiliary_loss_clip": 0.01066923,
      "auxiliary_loss_mlp": 0.00890918,
      "balance_loss_clip": 1.01213372,
      "balance_loss_mlp": 0.99999219,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7693518122744957,
      "language_loss": 0.63641477,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65599322,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 4.261692762374878
    },
    {
      "auxiliary_loss_clip": 0.01161268,
      "auxiliary_loss_mlp": 0.01023278,
      "balance_loss_clip": 1.04676056,
      "balance_loss_mlp": 1.01616716,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 2.2164836392805776,
      "language_loss": 0.72184992,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74369544,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.734312057495117
    },
    {
      "auxiliary_loss_clip": 0.01165181,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.04595494,
      "balance_loss_mlp": 1.02153027,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 1.9305993364145788,
      "language_loss": 0.69027293,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.7122159,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.5582711696624756
    },
    {
      "auxiliary_loss_clip": 0.01165454,
      "auxiliary_loss_mlp": 0.01021783,
      "balance_loss_clip": 1.04993176,
      "balance_loss_mlp": 1.01459503,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.6722846015427841,
      "language_loss": 0.77625686,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.7981292,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.6557023525238037
    },
    {
      "auxiliary_loss_clip": 0.01149999,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.04386389,
      "balance_loss_mlp": 1.01977253,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.2083550782259627,
      "language_loss": 0.81804335,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.8398236,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.6960158348083496
    },
    {
      "auxiliary_loss_clip": 0.01170397,
      "auxiliary_loss_mlp": 0.01019726,
      "balance_loss_clip": 1.04707623,
      "balance_loss_mlp": 1.01211417,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.20028149920873,
      "language_loss": 0.79683053,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81873178,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.5153627395629883
    },
    {
      "auxiliary_loss_clip": 0.01157211,
      "auxiliary_loss_mlp": 0.0090109,
      "balance_loss_clip": 1.0448252,
      "balance_loss_mlp": 1.0007391,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 1.8992026880208663,
      "language_loss": 0.62780803,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.64839101,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 2.642378568649292
    },
    {
      "auxiliary_loss_clip": 0.01141296,
      "auxiliary_loss_mlp": 0.01018929,
      "balance_loss_clip": 1.03907299,
      "balance_loss_mlp": 1.01122236,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 3.1818020209476825,
      "language_loss": 0.78793836,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80954057,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 3.5596628189086914
    },
    {
      "auxiliary_loss_clip": 0.0117408,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.05040765,
      "balance_loss_mlp": 1.01744628,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 5.953959735047187,
      "language_loss": 0.70615995,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72814971,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.5986084938049316
    },
    {
      "auxiliary_loss_clip": 0.01162158,
      "auxiliary_loss_mlp": 0.00901057,
      "balance_loss_clip": 1.04756963,
      "balance_loss_mlp": 1.00079918,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.3464387337389914,
      "language_loss": 0.71389449,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73452663,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.758392810821533
    },
    {
      "auxiliary_loss_clip": 0.01171514,
      "auxiliary_loss_mlp": 0.01023803,
      "balance_loss_clip": 1.0500685,
      "balance_loss_mlp": 1.01701081,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 2.487472793390084,
      "language_loss": 0.84507054,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86702371,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 3.5133137702941895
    },
    {
      "auxiliary_loss_clip": 0.01158353,
      "auxiliary_loss_mlp": 0.01021069,
      "balance_loss_clip": 1.04660988,
      "balance_loss_mlp": 1.01426554,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 2.11202348531344,
      "language_loss": 0.76795352,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.78974777,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.6415727138519287
    },
    {
      "auxiliary_loss_clip": 0.0116961,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.0494163,
      "balance_loss_mlp": 1.02077043,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 1.9214083733339835,
      "language_loss": 0.80158418,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.82355881,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.5249054431915283
    },
    {
      "auxiliary_loss_clip": 0.0115629,
      "auxiliary_loss_mlp": 0.01025943,
      "balance_loss_clip": 1.0471952,
      "balance_loss_mlp": 1.01877606,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 2.0036978287012883,
      "language_loss": 0.74139285,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.76321518,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.8033206462860107
    },
    {
      "auxiliary_loss_clip": 0.0115939,
      "auxiliary_loss_mlp": 0.0102205,
      "balance_loss_clip": 1.04636812,
      "balance_loss_mlp": 1.01487994,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.773885168057385,
      "language_loss": 0.7677452,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78955966,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.553657054901123
    },
    {
      "auxiliary_loss_clip": 0.01153337,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.04366994,
      "balance_loss_mlp": 1.0152626,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 3.14547183595729,
      "language_loss": 0.65757191,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.67933059,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.5858070850372314
    },
    {
      "auxiliary_loss_clip": 0.01164093,
      "auxiliary_loss_mlp": 0.01024102,
      "balance_loss_clip": 1.04736745,
      "balance_loss_mlp": 1.01696754,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.6887076423227154,
      "language_loss": 0.78226507,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80414701,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.5904898643493652
    },
    {
      "auxiliary_loss_clip": 0.01146752,
      "auxiliary_loss_mlp": 0.01027622,
      "balance_loss_clip": 1.04872918,
      "balance_loss_mlp": 1.02049351,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.7257634347001956,
      "language_loss": 0.78638309,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80812681,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 3.5859525203704834
    },
    {
      "auxiliary_loss_clip": 0.01166962,
      "auxiliary_loss_mlp": 0.01030501,
      "balance_loss_clip": 1.04874218,
      "balance_loss_mlp": 1.02297258,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.490604653268535,
      "language_loss": 0.79362071,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.81559527,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.572598934173584
    },
    {
      "auxiliary_loss_clip": 0.01147194,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04428875,
      "balance_loss_mlp": 1.01940799,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 2.048607285184609,
      "language_loss": 0.71066356,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.73240727,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.6112780570983887
    },
    {
      "auxiliary_loss_clip": 0.01154426,
      "auxiliary_loss_mlp": 0.01021207,
      "balance_loss_clip": 1.04548049,
      "balance_loss_mlp": 1.01349425,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 2.607128313412175,
      "language_loss": 0.75293338,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77468967,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.639216423034668
    },
    {
      "auxiliary_loss_clip": 0.01135898,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.04399014,
      "balance_loss_mlp": 1.01766658,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.8054339958046117,
      "language_loss": 0.77009809,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.79170835,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.672463893890381
    },
    {
      "auxiliary_loss_clip": 0.01173905,
      "auxiliary_loss_mlp": 0.01025134,
      "balance_loss_clip": 1.05143225,
      "balance_loss_mlp": 1.0175972,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.0246329315219573,
      "language_loss": 0.81157851,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83356887,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.586395740509033
    },
    {
      "auxiliary_loss_clip": 0.01150511,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.04272509,
      "balance_loss_mlp": 1.01597738,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.2935284863541248,
      "language_loss": 0.77265346,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79439104,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.5859687328338623
    },
    {
      "auxiliary_loss_clip": 0.01151134,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.04594827,
      "balance_loss_mlp": 1.01913977,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.7468413328492054,
      "language_loss": 0.80837488,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83014715,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.675769567489624
    },
    {
      "auxiliary_loss_clip": 0.01133343,
      "auxiliary_loss_mlp": 0.01024057,
      "balance_loss_clip": 1.04283929,
      "balance_loss_mlp": 1.01687431,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.9624856957349595,
      "language_loss": 0.78614092,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80771494,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.640503168106079
    },
    {
      "auxiliary_loss_clip": 0.01162362,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.04648113,
      "balance_loss_mlp": 1.017223,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.628514383135225,
      "language_loss": 0.73659223,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75845921,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 3.56013822555542
    },
    {
      "auxiliary_loss_clip": 0.01170503,
      "auxiliary_loss_mlp": 0.01026184,
      "balance_loss_clip": 1.05114722,
      "balance_loss_mlp": 1.01928186,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.7661187583113207,
      "language_loss": 0.75594848,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77791536,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.6370596885681152
    },
    {
      "auxiliary_loss_clip": 0.01128419,
      "auxiliary_loss_mlp": 0.01021788,
      "balance_loss_clip": 1.03923357,
      "balance_loss_mlp": 1.01450455,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.9498890675127947,
      "language_loss": 0.77049983,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79200196,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.6984570026397705
    },
    {
      "auxiliary_loss_clip": 0.01172064,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04809952,
      "balance_loss_mlp": 1.02002954,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.803114850440711,
      "language_loss": 0.88448548,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90648019,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.5943715572357178
    },
    {
      "auxiliary_loss_clip": 0.01066243,
      "auxiliary_loss_mlp": 0.01000798,
      "balance_loss_clip": 1.01066554,
      "balance_loss_mlp": 1.00001717,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7532483738841727,
      "language_loss": 0.62563467,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64630508,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.1271979808807373
    },
    {
      "auxiliary_loss_clip": 0.01145875,
      "auxiliary_loss_mlp": 0.01027796,
      "balance_loss_clip": 1.04614544,
      "balance_loss_mlp": 1.02056575,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 2.173710886655871,
      "language_loss": 0.67321444,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69495106,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.6318888664245605
    },
    {
      "auxiliary_loss_clip": 0.01160701,
      "auxiliary_loss_mlp": 0.01028064,
      "balance_loss_clip": 1.05037546,
      "balance_loss_mlp": 1.02042902,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.224889702479581,
      "language_loss": 0.73629737,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.75818503,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.5698087215423584
    },
    {
      "auxiliary_loss_clip": 0.01159993,
      "auxiliary_loss_mlp": 0.01022727,
      "balance_loss_clip": 1.0481652,
      "balance_loss_mlp": 1.01589608,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.5992673389236884,
      "language_loss": 0.7757653,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.79759252,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.5879015922546387
    },
    {
      "auxiliary_loss_clip": 0.01147512,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.04461837,
      "balance_loss_mlp": 1.01617265,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.7562127661465026,
      "language_loss": 0.75140619,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77311039,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 3.6196842193603516
    },
    {
      "auxiliary_loss_clip": 0.01140456,
      "auxiliary_loss_mlp": 0.01024909,
      "balance_loss_clip": 1.04375124,
      "balance_loss_mlp": 1.0173099,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.70631324657164,
      "language_loss": 0.8259241,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84757775,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.6135973930358887
    },
    {
      "auxiliary_loss_clip": 0.01070437,
      "auxiliary_loss_mlp": 0.01001235,
      "balance_loss_clip": 1.01100016,
      "balance_loss_mlp": 1.00046659,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8289029955620147,
      "language_loss": 0.58175957,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60247624,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.174974203109741
    },
    {
      "auxiliary_loss_clip": 0.01142029,
      "auxiliary_loss_mlp": 0.01028622,
      "balance_loss_clip": 1.04417348,
      "balance_loss_mlp": 1.02099574,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.470667974697879,
      "language_loss": 0.74330896,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76501548,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 3.600534677505493
    },
    {
      "auxiliary_loss_clip": 0.01162946,
      "auxiliary_loss_mlp": 0.00901128,
      "balance_loss_clip": 1.0457952,
      "balance_loss_mlp": 1.0006417,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.8515083448446734,
      "language_loss": 0.76752067,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78816146,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.680187225341797
    },
    {
      "auxiliary_loss_clip": 0.01172255,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.04739237,
      "balance_loss_mlp": 1.01938033,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.223002678720168,
      "language_loss": 0.70929706,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73128778,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6114494800567627
    },
    {
      "auxiliary_loss_clip": 0.01134039,
      "auxiliary_loss_mlp": 0.01020113,
      "balance_loss_clip": 1.0432024,
      "balance_loss_mlp": 1.01259649,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.5831872109534073,
      "language_loss": 0.80952621,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83106774,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.6997554302215576
    },
    {
      "auxiliary_loss_clip": 0.0116589,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.0491749,
      "balance_loss_mlp": 1.01534534,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 2.1423714972869807,
      "language_loss": 0.69112313,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71300781,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.736987590789795
    },
    {
      "auxiliary_loss_clip": 0.01161124,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.04757452,
      "balance_loss_mlp": 1.01685977,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 2.3883992308166366,
      "language_loss": 0.62566096,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64751649,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.6348233222961426
    },
    {
      "auxiliary_loss_clip": 0.01150533,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.04680598,
      "balance_loss_mlp": 1.01959515,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 1.9475147802053243,
      "language_loss": 0.61724007,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.63901502,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.568773031234741
    },
    {
      "auxiliary_loss_clip": 0.01154846,
      "auxiliary_loss_mlp": 0.01024118,
      "balance_loss_clip": 1.04759479,
      "balance_loss_mlp": 1.01658404,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.4789125827044156,
      "language_loss": 0.54122996,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.56301963,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 3.602987289428711
    },
    {
      "auxiliary_loss_clip": 0.01155673,
      "auxiliary_loss_mlp": 0.01022415,
      "balance_loss_clip": 1.04872108,
      "balance_loss_mlp": 1.01545668,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 3.910864047422222,
      "language_loss": 0.83482027,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85660124,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.654949903488159
    },
    {
      "auxiliary_loss_clip": 0.01167768,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.04727364,
      "balance_loss_mlp": 1.01771212,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 1.8827341778924853,
      "language_loss": 0.74605429,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.76799119,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.5948023796081543
    },
    {
      "auxiliary_loss_clip": 0.01104882,
      "auxiliary_loss_mlp": 0.01027035,
      "balance_loss_clip": 1.03777373,
      "balance_loss_mlp": 1.01960278,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 2.208874775576985,
      "language_loss": 0.73945343,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.76077259,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.82261323928833
    },
    {
      "auxiliary_loss_clip": 0.01158789,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.04757714,
      "balance_loss_mlp": 1.02033114,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 7.862218798148684,
      "language_loss": 0.74280477,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76467454,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.5669219493865967
    },
    {
      "auxiliary_loss_clip": 0.01134175,
      "auxiliary_loss_mlp": 0.01020458,
      "balance_loss_clip": 1.04091907,
      "balance_loss_mlp": 1.0130074,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.9235257102649344,
      "language_loss": 0.84194213,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86348844,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.78610897064209
    },
    {
      "auxiliary_loss_clip": 0.01168013,
      "auxiliary_loss_mlp": 0.01026181,
      "balance_loss_clip": 1.04717314,
      "balance_loss_mlp": 1.01806891,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.6985382212862177,
      "language_loss": 0.73403966,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75598156,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.5331549644470215
    },
    {
      "auxiliary_loss_clip": 0.01169416,
      "auxiliary_loss_mlp": 0.01029603,
      "balance_loss_clip": 1.04955816,
      "balance_loss_mlp": 1.0226028,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.87216176519664,
      "language_loss": 0.68241459,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70440483,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.550246000289917
    },
    {
      "auxiliary_loss_clip": 0.01146558,
      "auxiliary_loss_mlp": 0.01020348,
      "balance_loss_clip": 1.04426944,
      "balance_loss_mlp": 1.013309,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.6512677334342178,
      "language_loss": 0.80119628,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82286537,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.675567865371704
    },
    {
      "auxiliary_loss_clip": 0.01165699,
      "auxiliary_loss_mlp": 0.00901016,
      "balance_loss_clip": 1.04564142,
      "balance_loss_mlp": 1.0006988,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 2.495736270472003,
      "language_loss": 0.71553898,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73620617,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 3.547297239303589
    },
    {
      "auxiliary_loss_clip": 0.01153956,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.0450983,
      "balance_loss_mlp": 1.02214289,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.6592542522137224,
      "language_loss": 0.84771359,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.86955845,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.704335927963257
    },
    {
      "auxiliary_loss_clip": 0.01145367,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.04488611,
      "balance_loss_mlp": 1.02157903,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 3.068939272356181,
      "language_loss": 0.79635137,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81809771,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.5820398330688477
    },
    {
      "auxiliary_loss_clip": 0.01163072,
      "auxiliary_loss_mlp": 0.01025021,
      "balance_loss_clip": 1.04979908,
      "balance_loss_mlp": 1.01796377,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.751448484361704,
      "language_loss": 0.8000133,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82189429,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.7195005416870117
    },
    {
      "auxiliary_loss_clip": 0.01150666,
      "auxiliary_loss_mlp": 0.010254,
      "balance_loss_clip": 1.04589498,
      "balance_loss_mlp": 1.01831889,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 3.0193172608514205,
      "language_loss": 0.7928704,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.8146311,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.635906219482422
    },
    {
      "auxiliary_loss_clip": 0.01158591,
      "auxiliary_loss_mlp": 0.01021668,
      "balance_loss_clip": 1.04447651,
      "balance_loss_mlp": 1.01433671,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.3287491327735403,
      "language_loss": 0.65149724,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67329985,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.677506685256958
    },
    {
      "auxiliary_loss_clip": 0.01142818,
      "auxiliary_loss_mlp": 0.01023629,
      "balance_loss_clip": 1.04715693,
      "balance_loss_mlp": 1.01650047,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 2.8477055985427477,
      "language_loss": 0.87453449,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89619893,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.6464946269989014
    },
    {
      "auxiliary_loss_clip": 0.01169815,
      "auxiliary_loss_mlp": 0.01021183,
      "balance_loss_clip": 1.04680276,
      "balance_loss_mlp": 1.01388121,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.7149147178111077,
      "language_loss": 0.75323391,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77514386,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 3.585968255996704
    },
    {
      "auxiliary_loss_clip": 0.01162105,
      "auxiliary_loss_mlp": 0.01026307,
      "balance_loss_clip": 1.04832149,
      "balance_loss_mlp": 1.01853442,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.7983517794193076,
      "language_loss": 0.77729172,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.79917586,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.614506483078003
    },
    {
      "auxiliary_loss_clip": 0.01064005,
      "auxiliary_loss_mlp": 0.0100192,
      "balance_loss_clip": 1.01159358,
      "balance_loss_mlp": 1.00115156,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7571187190251425,
      "language_loss": 0.56517732,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58583653,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.108180046081543
    },
    {
      "auxiliary_loss_clip": 0.01157015,
      "auxiliary_loss_mlp": 0.01022061,
      "balance_loss_clip": 1.04786468,
      "balance_loss_mlp": 1.01488721,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 2.1642084427112898,
      "language_loss": 0.86784077,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.88963151,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.6340348720550537
    },
    {
      "auxiliary_loss_clip": 0.01175785,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.04893827,
      "balance_loss_mlp": 1.01784337,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 1.9603977674663304,
      "language_loss": 0.88617659,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.90819001,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 3.440854549407959
    },
    {
      "auxiliary_loss_clip": 0.01167918,
      "auxiliary_loss_mlp": 0.01030801,
      "balance_loss_clip": 1.04811072,
      "balance_loss_mlp": 1.02298665,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.8520731708269205,
      "language_loss": 0.76717627,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78916347,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.6439743041992188
    },
    {
      "auxiliary_loss_clip": 0.0113143,
      "auxiliary_loss_mlp": 0.01023633,
      "balance_loss_clip": 1.04172385,
      "balance_loss_mlp": 1.0163908,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.2896049066418116,
      "language_loss": 0.75293541,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77448601,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.7558512687683105
    },
    {
      "auxiliary_loss_clip": 0.01153332,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 1.05096078,
      "balance_loss_mlp": 1.01817226,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 2.4925091535223105,
      "language_loss": 0.87242067,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.89421117,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.6147632598876953
    },
    {
      "auxiliary_loss_clip": 0.01164191,
      "auxiliary_loss_mlp": 0.00901155,
      "balance_loss_clip": 1.04584169,
      "balance_loss_mlp": 1.00068426,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 2.0258071633918906,
      "language_loss": 0.67737544,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69802886,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.7371630668640137
    },
    {
      "auxiliary_loss_clip": 0.01139576,
      "auxiliary_loss_mlp": 0.01022734,
      "balance_loss_clip": 1.04026818,
      "balance_loss_mlp": 1.01521838,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 5.605871876799831,
      "language_loss": 0.76596713,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78759021,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.6026878356933594
    },
    {
      "auxiliary_loss_clip": 0.01156733,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04647732,
      "balance_loss_mlp": 1.02040553,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.3856934666188483,
      "language_loss": 0.69061148,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71245444,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.586456298828125
    },
    {
      "auxiliary_loss_clip": 0.01145558,
      "auxiliary_loss_mlp": 0.01022402,
      "balance_loss_clip": 1.0405364,
      "balance_loss_mlp": 1.01488554,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.0999149385403144,
      "language_loss": 0.80125403,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82293361,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 3.6020314693450928
    },
    {
      "auxiliary_loss_clip": 0.01169203,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.04772878,
      "balance_loss_mlp": 1.01647854,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 2.0431400212210127,
      "language_loss": 0.68293881,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70486909,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.636122465133667
    },
    {
      "auxiliary_loss_clip": 0.01131975,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.04611301,
      "balance_loss_mlp": 1.01996136,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.2383968756740624,
      "language_loss": 0.82342732,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.84502119,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.6928324699401855
    },
    {
      "auxiliary_loss_clip": 0.0115812,
      "auxiliary_loss_mlp": 0.01025545,
      "balance_loss_clip": 1.04778194,
      "balance_loss_mlp": 1.01835704,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 3.5650312513555824,
      "language_loss": 0.77986985,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80170649,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.638742685317993
    },
    {
      "auxiliary_loss_clip": 0.01142198,
      "auxiliary_loss_mlp": 0.01026835,
      "balance_loss_clip": 1.04271889,
      "balance_loss_mlp": 1.01967692,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 2.109077388815307,
      "language_loss": 0.92179471,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94348502,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.668586015701294
    },
    {
      "auxiliary_loss_clip": 0.01165548,
      "auxiliary_loss_mlp": 0.00902092,
      "balance_loss_clip": 1.04734838,
      "balance_loss_mlp": 1.00075245,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 3.855949802732656,
      "language_loss": 0.7835902,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80426657,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.629061698913574
    },
    {
      "auxiliary_loss_clip": 0.01171534,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.04778981,
      "balance_loss_mlp": 1.0184195,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 3.4549007663383757,
      "language_loss": 0.7821312,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.8041091,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 2.563372850418091
    },
    {
      "auxiliary_loss_clip": 0.011672,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.05049789,
      "balance_loss_mlp": 1.01936007,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.840346616593002,
      "language_loss": 0.828924,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.85086828,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.656118154525757
    },
    {
      "auxiliary_loss_clip": 0.01144042,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.04446745,
      "balance_loss_mlp": 1.02109289,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.2166780187176114,
      "language_loss": 0.77498698,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79671258,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 3.5335514545440674
    },
    {
      "auxiliary_loss_clip": 0.01172357,
      "auxiliary_loss_mlp": 0.01029624,
      "balance_loss_clip": 1.04844952,
      "balance_loss_mlp": 1.02166104,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 1.9813456503239062,
      "language_loss": 0.73623681,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.75825655,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.6033456325531006
    },
    {
      "auxiliary_loss_clip": 0.01148223,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.04780293,
      "balance_loss_mlp": 1.01937616,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 2.0112638831820244,
      "language_loss": 0.8330102,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85475719,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.6373002529144287
    },
    {
      "auxiliary_loss_clip": 0.01145948,
      "auxiliary_loss_mlp": 0.0102073,
      "balance_loss_clip": 1.04198837,
      "balance_loss_mlp": 1.01410151,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 2.917350819933069,
      "language_loss": 0.7140168,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73568356,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.6695406436920166
    },
    {
      "auxiliary_loss_clip": 0.01159284,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04590535,
      "balance_loss_mlp": 1.01523674,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.3048261114079764,
      "language_loss": 0.83438283,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85620809,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.610105037689209
    },
    {
      "auxiliary_loss_clip": 0.01170936,
      "auxiliary_loss_mlp": 0.01024143,
      "balance_loss_clip": 1.04825485,
      "balance_loss_mlp": 1.01687407,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.7293343664847463,
      "language_loss": 0.82469505,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84664583,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.667072296142578
    },
    {
      "auxiliary_loss_clip": 0.01169807,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.04758537,
      "balance_loss_mlp": 1.01704097,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 4.328905706212665,
      "language_loss": 0.75264359,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.77458858,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.689774513244629
    },
    {
      "auxiliary_loss_clip": 0.01169378,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04775858,
      "balance_loss_mlp": 1.01838732,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.5351882252506437,
      "language_loss": 0.78385043,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80579662,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.5720672607421875
    },
    {
      "auxiliary_loss_clip": 0.01160619,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.04545212,
      "balance_loss_mlp": 1.01815438,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 3.2611979210855058,
      "language_loss": 0.6923784,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71423751,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 3.5214314460754395
    },
    {
      "auxiliary_loss_clip": 0.01153317,
      "auxiliary_loss_mlp": 0.01024933,
      "balance_loss_clip": 1.04612446,
      "balance_loss_mlp": 1.01683831,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.830095835427832,
      "language_loss": 0.84294868,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86473119,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.65739369392395
    },
    {
      "auxiliary_loss_clip": 0.01164677,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.0506072,
      "balance_loss_mlp": 1.01971185,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 3.1328987924126004,
      "language_loss": 0.70601672,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72793531,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.5794074535369873
    },
    {
      "auxiliary_loss_clip": 0.01147661,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.04744303,
      "balance_loss_mlp": 1.02143717,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 2.0584305176906774,
      "language_loss": 0.74605691,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.76782918,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 3.5888097286224365
    },
    {
      "auxiliary_loss_clip": 0.01143428,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.04599297,
      "balance_loss_mlp": 1.02033436,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.089756369862771,
      "language_loss": 0.88972241,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91142374,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.6533496379852295
    },
    {
      "auxiliary_loss_clip": 0.01140627,
      "auxiliary_loss_mlp": 0.00901443,
      "balance_loss_clip": 1.04754448,
      "balance_loss_mlp": 1.0007782,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.6178103108708275,
      "language_loss": 0.72986674,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.75028741,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.7811708450317383
    },
    {
      "auxiliary_loss_clip": 0.01147345,
      "auxiliary_loss_mlp": 0.01022856,
      "balance_loss_clip": 1.04630291,
      "balance_loss_mlp": 1.01576948,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 2.8119680662629847,
      "language_loss": 0.70930296,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.73100495,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.5914158821105957
    },
    {
      "auxiliary_loss_clip": 0.01164057,
      "auxiliary_loss_mlp": 0.00901422,
      "balance_loss_clip": 1.04662895,
      "balance_loss_mlp": 1.00080824,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.841645800550299,
      "language_loss": 0.84379977,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86445457,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.7432150840759277
    },
    {
      "auxiliary_loss_clip": 0.01149444,
      "auxiliary_loss_mlp": 0.01025081,
      "balance_loss_clip": 1.0428158,
      "balance_loss_mlp": 1.01740694,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.867601417435578,
      "language_loss": 0.79125714,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81300235,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.6257331371307373
    },
    {
      "auxiliary_loss_clip": 0.0114591,
      "auxiliary_loss_mlp": 0.01022219,
      "balance_loss_clip": 1.04166985,
      "balance_loss_mlp": 1.0140357,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 1.9591720385800844,
      "language_loss": 0.68295008,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70463133,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.7088518142700195
    },
    {
      "auxiliary_loss_clip": 0.01147245,
      "auxiliary_loss_mlp": 0.01026993,
      "balance_loss_clip": 1.04308963,
      "balance_loss_mlp": 1.01869607,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.1801223856223753,
      "language_loss": 0.79671592,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81845832,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 3.6029486656188965
    },
    {
      "auxiliary_loss_clip": 0.01162399,
      "auxiliary_loss_mlp": 0.01025842,
      "balance_loss_clip": 1.04719174,
      "balance_loss_mlp": 1.01851928,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.962872303848381,
      "language_loss": 0.74856055,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77044296,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.6007068157196045
    },
    {
      "auxiliary_loss_clip": 0.01150506,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.04699326,
      "balance_loss_mlp": 1.02372909,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 2.103912249993597,
      "language_loss": 0.73576975,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.7575863,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.6590559482574463
    },
    {
      "auxiliary_loss_clip": 0.01152477,
      "auxiliary_loss_mlp": 0.01023838,
      "balance_loss_clip": 1.04629922,
      "balance_loss_mlp": 1.01668596,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 2.9950253217979377,
      "language_loss": 0.68204778,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70381093,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.6258771419525146
    },
    {
      "auxiliary_loss_clip": 0.01170911,
      "auxiliary_loss_mlp": 0.01021696,
      "balance_loss_clip": 1.04951882,
      "balance_loss_mlp": 1.01413202,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 2.015426388214751,
      "language_loss": 0.71948051,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.74140656,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.5989794731140137
    },
    {
      "auxiliary_loss_clip": 0.01155647,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.04495549,
      "balance_loss_mlp": 1.02304626,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.0703950670220013,
      "language_loss": 0.85664749,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87850946,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.5867209434509277
    },
    {
      "auxiliary_loss_clip": 0.01067017,
      "auxiliary_loss_mlp": 0.01001794,
      "balance_loss_clip": 1.01085281,
      "balance_loss_mlp": 1.00098288,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7245359917983971,
      "language_loss": 0.59636068,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61704886,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.2752277851104736
    },
    {
      "auxiliary_loss_clip": 0.01102898,
      "auxiliary_loss_mlp": 0.01022164,
      "balance_loss_clip": 1.035887,
      "balance_loss_mlp": 1.01495755,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.4762894603863415,
      "language_loss": 0.57396019,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59521079,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 3.073755979537964
    },
    {
      "auxiliary_loss_clip": 0.01061225,
      "auxiliary_loss_mlp": 0.01001784,
      "balance_loss_clip": 1.01124215,
      "balance_loss_mlp": 1.00097966,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7997967195782854,
      "language_loss": 0.57746005,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59809017,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.3779633045196533
    },
    {
      "auxiliary_loss_clip": 0.01149758,
      "auxiliary_loss_mlp": 0.01021269,
      "balance_loss_clip": 1.04584527,
      "balance_loss_mlp": 1.01455998,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 2.943407983769285,
      "language_loss": 0.74889642,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.7706067,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 3.8603718280792236
    },
    {
      "auxiliary_loss_clip": 0.01150468,
      "auxiliary_loss_mlp": 0.01024629,
      "balance_loss_clip": 1.0457027,
      "balance_loss_mlp": 1.01732445,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.8335940090498248,
      "language_loss": 0.58495831,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.60670924,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.7461719512939453
    },
    {
      "auxiliary_loss_clip": 0.01152048,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.04617214,
      "balance_loss_mlp": 1.01717389,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 2.5241841288524522,
      "language_loss": 0.68541783,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70718384,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.645399332046509
    },
    {
      "auxiliary_loss_clip": 0.01153014,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04355812,
      "balance_loss_mlp": 1.01899326,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 1.8876025209299399,
      "language_loss": 0.81500989,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.83679771,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.661587715148926
    },
    {
      "auxiliary_loss_clip": 0.01149843,
      "auxiliary_loss_mlp": 0.01023934,
      "balance_loss_clip": 1.044438,
      "balance_loss_mlp": 1.01587558,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 3.468858539718621,
      "language_loss": 0.82142305,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.84316081,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.6817526817321777
    },
    {
      "auxiliary_loss_clip": 0.01128689,
      "auxiliary_loss_mlp": 0.01029065,
      "balance_loss_clip": 1.04022431,
      "balance_loss_mlp": 1.02126861,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.022365615204012,
      "language_loss": 0.67168719,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.69326472,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.7569375038146973
    },
    {
      "auxiliary_loss_clip": 0.01151996,
      "auxiliary_loss_mlp": 0.01024772,
      "balance_loss_clip": 1.04632151,
      "balance_loss_mlp": 1.01747024,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.2013249891654163,
      "language_loss": 0.7286433,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.75041103,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.662842273712158
    },
    {
      "auxiliary_loss_clip": 0.01061011,
      "auxiliary_loss_mlp": 0.01001046,
      "balance_loss_clip": 1.0101409,
      "balance_loss_mlp": 1.00021732,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6882934723810334,
      "language_loss": 0.55677152,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.5773921,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 4.147134304046631
    },
    {
      "auxiliary_loss_clip": 0.01151437,
      "auxiliary_loss_mlp": 0.01020496,
      "balance_loss_clip": 1.0463841,
      "balance_loss_mlp": 1.0139277,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.828540109782812,
      "language_loss": 0.75692379,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.77864313,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.6732327938079834
    },
    {
      "auxiliary_loss_clip": 0.01133314,
      "auxiliary_loss_mlp": 0.00901872,
      "balance_loss_clip": 1.04051006,
      "balance_loss_mlp": 1.00079203,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.0267986714045056,
      "language_loss": 0.75429451,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77464628,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.759586811065674
    },
    {
      "auxiliary_loss_clip": 0.01159468,
      "auxiliary_loss_mlp": 0.00900337,
      "balance_loss_clip": 1.04908299,
      "balance_loss_mlp": 1.00080633,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.9841323966891915,
      "language_loss": 0.74600267,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.76660073,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.656942844390869
    },
    {
      "auxiliary_loss_clip": 0.01134467,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04417884,
      "balance_loss_mlp": 1.01853585,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 2.654209324522488,
      "language_loss": 0.80000901,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.82161891,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 3.7596704959869385
    },
    {
      "auxiliary_loss_clip": 0.01164675,
      "auxiliary_loss_mlp": 0.01032611,
      "balance_loss_clip": 1.04687548,
      "balance_loss_mlp": 1.02523839,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.8950908031157823,
      "language_loss": 0.74766546,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.7696383,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.635504961013794
    },
    {
      "auxiliary_loss_clip": 0.01167065,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.05025172,
      "balance_loss_mlp": 1.01912546,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 4.586536795127082,
      "language_loss": 0.77634209,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79827768,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.598660707473755
    },
    {
      "auxiliary_loss_clip": 0.01172532,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.04875338,
      "balance_loss_mlp": 1.02096057,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.5159503764318276,
      "language_loss": 0.76806271,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.79007518,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.537019729614258
    },
    {
      "auxiliary_loss_clip": 0.01133701,
      "auxiliary_loss_mlp": 0.01026721,
      "balance_loss_clip": 1.04696774,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.4827543977716067,
      "language_loss": 0.6764729,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69807708,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.708388090133667
    },
    {
      "auxiliary_loss_clip": 0.01160251,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.04675138,
      "balance_loss_mlp": 1.01739478,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.8209488450194633,
      "language_loss": 0.68658119,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.7084341,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.577040910720825
    },
    {
      "auxiliary_loss_clip": 0.01060754,
      "auxiliary_loss_mlp": 0.01002082,
      "balance_loss_clip": 1.0135181,
      "balance_loss_mlp": 1.00113475,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8800238333532007,
      "language_loss": 0.61071599,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63134444,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 4.130373954772949
    },
    {
      "auxiliary_loss_clip": 0.01170279,
      "auxiliary_loss_mlp": 0.01022759,
      "balance_loss_clip": 1.04861116,
      "balance_loss_mlp": 1.01581264,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 3.8689160276372863,
      "language_loss": 0.83581531,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85774565,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.56036114692688
    },
    {
      "auxiliary_loss_clip": 0.01143702,
      "auxiliary_loss_mlp": 0.01027983,
      "balance_loss_clip": 1.04203153,
      "balance_loss_mlp": 1.02046657,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 2.2123895094321107,
      "language_loss": 0.80857277,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.8302896,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.6455254554748535
    },
    {
      "auxiliary_loss_clip": 0.01161093,
      "auxiliary_loss_mlp": 0.01020044,
      "balance_loss_clip": 1.04519963,
      "balance_loss_mlp": 1.01274848,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 3.0307086719243235,
      "language_loss": 0.86176157,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88357294,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 2.595710515975952
    },
    {
      "auxiliary_loss_clip": 0.0115094,
      "auxiliary_loss_mlp": 0.01024394,
      "balance_loss_clip": 1.04387808,
      "balance_loss_mlp": 1.01695514,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.3768662335555937,
      "language_loss": 0.82932395,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.85107732,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.6046226024627686
    },
    {
      "auxiliary_loss_clip": 0.01155812,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.04596841,
      "balance_loss_mlp": 1.01694679,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 2.991137339100003,
      "language_loss": 0.83275199,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85455263,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.588468313217163
    },
    {
      "auxiliary_loss_clip": 0.01158277,
      "auxiliary_loss_mlp": 0.00901233,
      "balance_loss_clip": 1.04707241,
      "balance_loss_mlp": 1.00075412,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 2.0883043827543863,
      "language_loss": 0.81072062,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.8313157,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.5995066165924072
    },
    {
      "auxiliary_loss_clip": 0.0116787,
      "auxiliary_loss_mlp": 0.01030282,
      "balance_loss_clip": 1.04881334,
      "balance_loss_mlp": 1.02248597,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.1616319263525545,
      "language_loss": 0.71084034,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73282182,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.655613422393799
    },
    {
      "auxiliary_loss_clip": 0.01175572,
      "auxiliary_loss_mlp": 0.01025728,
      "balance_loss_clip": 1.050143,
      "balance_loss_mlp": 1.01799095,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 3.186773733427749,
      "language_loss": 0.6145094,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63652241,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.52134370803833
    },
    {
      "auxiliary_loss_clip": 0.01169015,
      "auxiliary_loss_mlp": 0.01026002,
      "balance_loss_clip": 1.04816484,
      "balance_loss_mlp": 1.01883113,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 3.426874040460278,
      "language_loss": 0.5583145,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.58026469,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.556457042694092
    },
    {
      "auxiliary_loss_clip": 0.0116916,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 1.04950464,
      "balance_loss_mlp": 1.01864696,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.3176551725012446,
      "language_loss": 0.7941041,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.816064,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 3.530959367752075
    },
    {
      "auxiliary_loss_clip": 0.01152688,
      "auxiliary_loss_mlp": 0.00901449,
      "balance_loss_clip": 1.046929,
      "balance_loss_mlp": 1.00087631,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 1.884121983961511,
      "language_loss": 0.62908751,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.64962888,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.660095691680908
    },
    {
      "auxiliary_loss_clip": 0.01138001,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.0425483,
      "balance_loss_mlp": 1.02216232,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 4.8714284090975575,
      "language_loss": 0.85856998,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88024366,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.7531373500823975
    },
    {
      "auxiliary_loss_clip": 0.01143309,
      "auxiliary_loss_mlp": 0.01020927,
      "balance_loss_clip": 1.04805493,
      "balance_loss_mlp": 1.01423025,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.7747991661455904,
      "language_loss": 0.72343463,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74507701,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.6978728771209717
    },
    {
      "auxiliary_loss_clip": 0.01151879,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.04660773,
      "balance_loss_mlp": 1.02353382,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 4.103235142465084,
      "language_loss": 0.76724505,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78907079,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.661580801010132
    },
    {
      "auxiliary_loss_clip": 0.01151835,
      "auxiliary_loss_mlp": 0.01021504,
      "balance_loss_clip": 1.04553294,
      "balance_loss_mlp": 1.0148133,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 1.8065050626099242,
      "language_loss": 0.75383103,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77556443,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.716143846511841
    },
    {
      "auxiliary_loss_clip": 0.01130319,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.04262102,
      "balance_loss_mlp": 1.01677036,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.1531236859704306,
      "language_loss": 0.71186274,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73341572,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.7216970920562744
    },
    {
      "auxiliary_loss_clip": 0.01070428,
      "auxiliary_loss_mlp": 0.01000946,
      "balance_loss_clip": 1.01082516,
      "balance_loss_mlp": 1.0001359,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7920433431105377,
      "language_loss": 0.5575242,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57823789,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 4.145308494567871
    },
    {
      "auxiliary_loss_clip": 0.01155147,
      "auxiliary_loss_mlp": 0.01022188,
      "balance_loss_clip": 1.04904211,
      "balance_loss_mlp": 1.01491594,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 3.026240270500245,
      "language_loss": 0.69441557,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71618885,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.6542320251464844
    },
    {
      "auxiliary_loss_clip": 0.01140004,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.04425919,
      "balance_loss_mlp": 1.01835191,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.724952148987012,
      "language_loss": 0.70809352,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.72975326,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.7283334732055664
    },
    {
      "auxiliary_loss_clip": 0.01158514,
      "auxiliary_loss_mlp": 0.01023735,
      "balance_loss_clip": 1.04543638,
      "balance_loss_mlp": 1.01618958,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.0410398498520297,
      "language_loss": 0.84449291,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86631536,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.587385892868042
    },
    {
      "auxiliary_loss_clip": 0.01157812,
      "auxiliary_loss_mlp": 0.0102661,
      "balance_loss_clip": 1.0485667,
      "balance_loss_mlp": 1.01927257,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.1844090717521203,
      "language_loss": 0.71341789,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73526216,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 3.5801093578338623
    },
    {
      "auxiliary_loss_clip": 0.01160745,
      "auxiliary_loss_mlp": 0.01029026,
      "balance_loss_clip": 1.0459286,
      "balance_loss_mlp": 1.02164698,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.794099415734497,
      "language_loss": 0.62985396,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65175164,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.683943510055542
    },
    {
      "auxiliary_loss_clip": 0.01160184,
      "auxiliary_loss_mlp": 0.01021125,
      "balance_loss_clip": 1.04743564,
      "balance_loss_mlp": 1.01403189,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 1.975191695900406,
      "language_loss": 0.87226766,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89408076,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.582688808441162
    },
    {
      "auxiliary_loss_clip": 0.01156412,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04352045,
      "balance_loss_mlp": 1.01674247,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 2.1675071075506573,
      "language_loss": 0.77840102,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.80020988,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.649454116821289
    },
    {
      "auxiliary_loss_clip": 0.01172058,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.04744995,
      "balance_loss_mlp": 1.0197494,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 2.757782477619303,
      "language_loss": 0.71998811,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74198955,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.5213611125946045
    },
    {
      "auxiliary_loss_clip": 0.01142556,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.04648316,
      "balance_loss_mlp": 1.02012241,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 3.000463763490526,
      "language_loss": 0.73575771,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75745839,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.8053383827209473
    },
    {
      "auxiliary_loss_clip": 0.01134751,
      "auxiliary_loss_mlp": 0.01025659,
      "balance_loss_clip": 1.03948474,
      "balance_loss_mlp": 1.01795816,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.6098393833373899,
      "language_loss": 0.79899657,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82060063,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 3.7330636978149414
    },
    {
      "auxiliary_loss_clip": 0.01140694,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.04551768,
      "balance_loss_mlp": 1.0236733,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 1.9820382735458333,
      "language_loss": 0.7933405,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81507844,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.722287654876709
    },
    {
      "auxiliary_loss_clip": 0.01070521,
      "auxiliary_loss_mlp": 0.01001266,
      "balance_loss_clip": 1.01088643,
      "balance_loss_mlp": 1.00046182,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8258462326901239,
      "language_loss": 0.62840253,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64912045,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.0366134643554688
    },
    {
      "auxiliary_loss_clip": 0.01153069,
      "auxiliary_loss_mlp": 0.01029673,
      "balance_loss_clip": 1.05080569,
      "balance_loss_mlp": 1.0224545,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 5.042452264045632,
      "language_loss": 0.69021142,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.71203887,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.659841537475586
    },
    {
      "auxiliary_loss_clip": 0.01154016,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.04472947,
      "balance_loss_mlp": 1.01871777,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.036519460461662,
      "language_loss": 0.88336957,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90517068,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.6092936992645264
    },
    {
      "auxiliary_loss_clip": 0.0106267,
      "auxiliary_loss_mlp": 0.01001682,
      "balance_loss_clip": 1.01130617,
      "balance_loss_mlp": 1.00088334,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7531327835102664,
      "language_loss": 0.57693923,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.5975827,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.1965863704681396
    },
    {
      "auxiliary_loss_clip": 0.01171869,
      "auxiliary_loss_mlp": 0.01025772,
      "balance_loss_clip": 1.04822767,
      "balance_loss_mlp": 1.01777887,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.9620521171387895,
      "language_loss": 0.72821951,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.75019598,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.601696252822876
    },
    {
      "auxiliary_loss_clip": 0.01150007,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.04756021,
      "balance_loss_mlp": 1.02276945,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 2.627138339397959,
      "language_loss": 0.6667614,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.6885671,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.6872832775115967
    },
    {
      "auxiliary_loss_clip": 0.01159008,
      "auxiliary_loss_mlp": 0.01027981,
      "balance_loss_clip": 1.04699159,
      "balance_loss_mlp": 1.02034569,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.294196312735164,
      "language_loss": 0.73622638,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75809634,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.6654562950134277
    },
    {
      "auxiliary_loss_clip": 0.01152208,
      "auxiliary_loss_mlp": 0.01023643,
      "balance_loss_clip": 1.04488707,
      "balance_loss_mlp": 1.01602006,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.850012317163128,
      "language_loss": 0.79436529,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81612372,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 3.519956111907959
    },
    {
      "auxiliary_loss_clip": 0.01120651,
      "auxiliary_loss_mlp": 0.01024559,
      "balance_loss_clip": 1.04123163,
      "balance_loss_mlp": 1.01690578,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.899604902657872,
      "language_loss": 0.63074481,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65219688,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.805203676223755
    },
    {
      "auxiliary_loss_clip": 0.01129948,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.03989148,
      "balance_loss_mlp": 1.01625097,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 8.193204014947439,
      "language_loss": 0.78301603,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80455804,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.658141613006592
    },
    {
      "auxiliary_loss_clip": 0.0113968,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 1.04574203,
      "balance_loss_mlp": 1.01806736,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.846318625210459,
      "language_loss": 0.8378458,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.8594979,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.7155544757843018
    },
    {
      "auxiliary_loss_clip": 0.01168209,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.05098438,
      "balance_loss_mlp": 1.01805019,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 3.949074268818397,
      "language_loss": 0.7820313,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80397195,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.571570634841919
    },
    {
      "auxiliary_loss_clip": 0.01151102,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04274869,
      "balance_loss_mlp": 1.02065134,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.680821432422887,
      "language_loss": 0.73123705,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75303388,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.701730489730835
    },
    {
      "auxiliary_loss_clip": 0.01150553,
      "auxiliary_loss_mlp": 0.01020401,
      "balance_loss_clip": 1.04624367,
      "balance_loss_mlp": 1.01325488,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 1.9678981479413151,
      "language_loss": 0.73875922,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76046872,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.845839738845825
    },
    {
      "auxiliary_loss_clip": 0.01154414,
      "auxiliary_loss_mlp": 0.01024572,
      "balance_loss_clip": 1.04836559,
      "balance_loss_mlp": 1.01740432,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 1.9173757759382652,
      "language_loss": 0.79074049,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81253034,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.6694817543029785
    },
    {
      "auxiliary_loss_clip": 0.01171723,
      "auxiliary_loss_mlp": 0.01023359,
      "balance_loss_clip": 1.04783058,
      "balance_loss_mlp": 1.01627529,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 1.7725675820728715,
      "language_loss": 0.74605256,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.7680034,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 3.548870325088501
    },
    {
      "auxiliary_loss_clip": 0.01138183,
      "auxiliary_loss_mlp": 0.01023323,
      "balance_loss_clip": 1.04170704,
      "balance_loss_mlp": 1.0163914,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.4952145073071836,
      "language_loss": 0.76280379,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.78441888,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.6441478729248047
    },
    {
      "auxiliary_loss_clip": 0.01149909,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.04670489,
      "balance_loss_mlp": 1.02066612,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 2.382033920367856,
      "language_loss": 0.62186259,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64364195,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.8961737155914307
    },
    {
      "auxiliary_loss_clip": 0.01168089,
      "auxiliary_loss_mlp": 0.01019321,
      "balance_loss_clip": 1.04613936,
      "balance_loss_mlp": 1.01191258,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 5.214392862488196,
      "language_loss": 0.7064454,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72831953,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 3.50484299659729
    },
    {
      "auxiliary_loss_clip": 0.01161768,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.0460515,
      "balance_loss_mlp": 1.01731801,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.135501875938079,
      "language_loss": 0.80957973,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83144999,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.664193630218506
    },
    {
      "auxiliary_loss_clip": 0.01148992,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 1.04653418,
      "balance_loss_mlp": 1.02192485,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 1.8630602594530585,
      "language_loss": 0.63667834,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.65846789,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.656123161315918
    },
    {
      "auxiliary_loss_clip": 0.01142862,
      "auxiliary_loss_mlp": 0.01029,
      "balance_loss_clip": 1.04601848,
      "balance_loss_mlp": 1.02172804,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 2.903957348444643,
      "language_loss": 0.77077127,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79248989,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.00789475440979
    },
    {
      "auxiliary_loss_clip": 0.01116378,
      "auxiliary_loss_mlp": 0.01015135,
      "balance_loss_clip": 1.02770138,
      "balance_loss_mlp": 1.01091552,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.126778700320345,
      "language_loss": 0.55016589,
      "learning_rate": 0.0,
      "loss": 0.57148105,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.154669761657715
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.790059989500126,
      "train_runtime": 24784.9939,
      "train_samples_per_second": 13.421,
      "train_steps_per_second": 0.336
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 8316,
  "num_input_tokens_seen": 179513955,
  "num_train_epochs": 1,
  "save_steps": 1664,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.996749092776837e+17,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}